"""
We use following lines because we are running on Google Colab
If you are running notebook on a local computer, you don't need this cell
"""
from google.colab import drive
drive.mount('/content/gdrive')
import os
os.chdir('/content/gdrive/My Drive/finch/tensorflow1/semantic_parsing/tree_slu/main')
%tensorflow_version 1.x
!pip install tensorflow-hub
!pip install texar
Requirement already satisfied: tensorflow-hub in /usr/local/lib/python3.6/dist-packages (0.7.0)
Requirement already satisfied: protobuf>=3.4.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-hub) (3.10.0)
Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-hub) (1.12.0)
Requirement already satisfied: numpy>=1.12.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-hub) (1.17.4)
Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from protobuf>=3.4.0->tensorflow-hub) (41.4.0)
Collecting texar
Downloading https://files.pythonhosted.org/packages/f4/a7/c439f300fa7bc53a798e48da4896a62aa92053f93e78c160d74b6742531b/texar-0.2.3.tar.gz (260kB)
|████████████████████████████████| 266kB 9.1MB/s
Collecting numpy<1.17.0
Downloading https://files.pythonhosted.org/packages/98/87/41283370f942f647422581eed16df4b653a744a3e9d5cfbb9aee0440f6eb/numpy-1.16.5-cp36-cp36m-manylinux1_x86_64.whl (17.4MB)
|████████████████████████████████| 17.4MB 476kB/s
Requirement already satisfied: pathlib>=1.0 in /usr/local/lib/python3.6/dist-packages (from texar) (1.0.1)
Requirement already satisfied: pyyaml in /usr/local/lib/python3.6/dist-packages (from texar) (3.13)
Requirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from texar) (2.21.0)
Collecting funcsigs>=1.0.2
Downloading https://files.pythonhosted.org/packages/69/cb/f5be453359271714c01b9bd06126eaf2e368f1fddfff30818754b5ac2328/funcsigs-1.0.2-py2.py3-none-any.whl
Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from texar) (19.2)
Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->texar) (2019.9.11)
Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->texar) (3.0.4)
Requirement already satisfied: idna<2.9,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->texar) (2.8)
Requirement already satisfied: urllib3<1.25,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests->texar) (1.24.3)
Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from packaging->texar) (2.4.5)
Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from packaging->texar) (1.12.0)
Building wheels for collected packages: texar
Building wheel for texar (setup.py) ... done
Created wheel for texar: filename=texar-0.2.3-cp36-none-any.whl size=395469 sha256=72dcb299e89a8ce749a5a6e0bd527b2739205b8d6ad213b8f1f92aa96ab6fb76
Stored in directory: /root/.cache/pip/wheels/a9/00/3d/b2c9a2c98e2102bb3fe24ebaa5bd09303fa392aab6f77d78e8
Successfully built texar
ERROR: datascience 0.10.6 has requirement folium==0.2.1, but you'll have folium 0.8.3 which is incompatible.
ERROR: albumentations 0.1.12 has requirement imgaug<0.2.7,>=0.2.5, but you'll have imgaug 0.2.9 which is incompatible.
Installing collected packages: numpy, funcsigs, texar
Found existing installation: numpy 1.17.4
Uninstalling numpy-1.17.4:
Successfully uninstalled numpy-1.17.4
Successfully installed funcsigs-1.0.2 numpy-1.16.5 texar-0.2.3
import tensorflow as tf
import tensorflow_hub as hub
import texar.tf as tx
import numpy as np
import pprint
import logging
import time
import nltk
from texar.tf.modules import TransformerDecoder
from pathlib import Path
print("TensorFlow Version", tf.__version__)
print('GPU Enabled:', tf.test.is_gpu_available())
TensorFlow Version 1.15.0 GPU Enabled: True
# stream data from text files
def data_generator(f_path, params):
with open(f_path) as f:
print('Reading', f_path)
for line in f:
text_raw, text_tokenized, label = line.split('\t')
text_tokenized = text_tokenized.lower().split()
label = label.replace('[', '[ ').lower().split()
#source = [params['tgt2idx'].get(w, len(params['tgt2idx'])) for w in text_tokenized]
target = [params['tgt2idx'].get(w, len(params['tgt2idx'])) for w in label]
target_in = [1] + target
target_out = target + [2]
yield (text_tokenized, (target_in, target_out))
def dataset(is_training, params):
_shapes = ([None], ([None], [None]))
_types = (tf.string, (tf.int32, tf.int32))
_pads = ('<pad>', (0, 0))
if is_training:
ds = tf.data.Dataset.from_generator(
lambda: data_generator(params['train_path'], params),
output_shapes = _shapes,
output_types = _types,)
ds = ds.shuffle(params['buffer_size'])
ds = ds.padded_batch(params['batch_size'], _shapes, _pads)
ds = ds.repeat(10)
ds = ds.prefetch(tf.data.experimental.AUTOTUNE)
else:
ds = tf.data.Dataset.from_generator(
lambda: data_generator(params['test_path'], params),
output_shapes = _shapes,
output_types = _types,)
ds = ds.padded_batch(4*params['batch_size'], _shapes, _pads)
ds = ds.prefetch(tf.data.experimental.AUTOTUNE)
return ds
def forward(raw_texts, labels, mode):
vocab = tf.contrib.lookup.index_table_from_file(
params['vocab_tgt_path'], num_oov_buckets=1)
words = vocab.lookup(raw_texts)
words_len = tf.count_nonzero(words, 1, dtype=tf.int32)
is_training = (mode == tf.estimator.ModeKeys.TRAIN)
batch_sz = tf.shape(words)[0]
with tf.variable_scope('Embedding'):
elmo = hub.Module("https://tfhub.dev/google/elmo/2", trainable=False)
e = elmo(inputs={'tokens':raw_texts, 'sequence_len':words_len,}, signature="tokens", as_dict=True)['lstm_outputs1']
e = tf.layers.dropout(e, params['dropout_rate'], training=is_training)
e = tf.layers.dense(e, params['rnn_units'], params['activation'])
embedding = tf.Variable(np.load('../vocab/word.npy'),
dtype=tf.float32,
name='glove')
x = tf.nn.embedding_lookup(embedding, words)
x = tf.concat((x, e), -1)
x = tf.layers.dropout(x, params['dropout_rate'], training=is_training)
with tf.variable_scope('Encoder'):
t = tf.transpose(x, perm=[1, 0, 2]) # Need time-major
lstm_cell_fw = tf.contrib.rnn.LSTMBlockFusedCell(params['rnn_units'])
lstm_cell_bw = tf.contrib.rnn.LSTMBlockFusedCell(params['rnn_units'])
lstm_cell_bw = tf.contrib.rnn.TimeReversedFusedRNN(lstm_cell_bw)
o_fw, s_fw = lstm_cell_fw(t, dtype=tf.float32, sequence_length=words_len)
o_bw, s_bw = lstm_cell_bw(t, dtype=tf.float32, sequence_length=words_len)
enc_out = tf.concat([o_fw, o_bw], axis=-1)
enc_out = tf.transpose(enc_out, perm=[1, 0, 2])
embedding = tf.concat([tf.zeros(shape=[1, 300]), embedding[1:, :]], axis=0)
pos_embedder = tx.modules.SinusoidsPositionEmbedder(
position_size = params['max_dec_len'],
hparams = config_model.position_embedder_hparams)
with tf.variable_scope('Decoder'):
decoder = TransformerDecoder(vocab_size=len(params['tgt2idx'])+1,
output_layer=tf.transpose(embedding, (1, 0)),
hparams=config_model.decoder)
if is_training or (mode == tf.estimator.ModeKeys.EVAL):
dec_inputs, dec_outputs = labels
dec_seq_len = tf.count_nonzero(dec_inputs, 1, dtype=tf.int32)
dec_inputs = tf.nn.embedding_lookup(embedding, dec_inputs)
dec_inputs = (dec_inputs * config_model.hidden_dim ** 0.5) + pos_embedder(sequence_length=dec_seq_len)
#dec_inputs = tf.layers.dropout(dec_inputs, params['dropout_rate'], training=is_training)
outputs = decoder(
memory=enc_out,
memory_sequence_length=words_len,
inputs=dec_inputs,
decoding_strategy='train_greedy',
mode=tf.estimator.ModeKeys.TRAIN
)
return outputs.logits
else:
start_tokens = tf.fill([batch_sz], 1)
def _embedding_fn(x, y):
x_w_embed = tf.nn.embedding_lookup(embedding, x)
y_p_embed = pos_embedder(y)
return x_w_embed * config_model.hidden_dim ** 0.5 + y_p_embed
predictions = decoder(
memory=enc_out,
memory_sequence_length=words_len,
beam_width=params['beam_width'],
length_penalty=params['length_penalty'],
start_tokens=start_tokens,
end_token=2,
embedding=_embedding_fn,
max_decoding_length=params['max_dec_len'],
mode=tf.estimator.ModeKeys.PREDICT)
return predictions['sample_id'][:, :, 0]
def clip_grads(loss):
variables = tf.trainable_variables()
pprint.pprint(variables)
grads = tf.gradients(loss, variables)
clipped_grads, _ = tf.clip_by_global_norm(grads, params['clip_norm'])
return zip(clipped_grads, variables)
def model_fn(features, labels, mode, params):
logits_or_ids = forward(features, labels, mode)
if mode == tf.estimator.ModeKeys.PREDICT:
return tf.estimator.EstimatorSpec(mode, predictions=logits_or_ids)
dec_inputs, dec_outputs = labels
loss_op = tf.losses.softmax_cross_entropy(onehot_labels = tf.one_hot(dec_outputs, len(params['tgt2idx'])+1),
logits = logits_or_ids,
weights = tf.to_float(tf.sign(dec_outputs)),
label_smoothing = .1)
if mode == tf.estimator.ModeKeys.TRAIN:
global_step=tf.train.get_or_create_global_step()
lr = tf.train.exponential_decay(
params['lr'], global_step, 1000, .99)
"""
lr = (2 * (params['rnn_units'] ** -0.5)) \
* tf.math.minimum(1.0, tf.to_float(global_step / params['warmup_steps'])) \
* (1 / tf.math.sqrt(tf.to_float(tf.math.maximum(global_step, params['warmup_steps']))))
"""
train_op = tf.train.AdamOptimizer(lr).apply_gradients(
clip_grads(loss_op), global_step=global_step)
hook = tf.train.LoggingTensorHook({'lr': lr}, every_n_iter=100)
return tf.estimator.EstimatorSpec(
mode=mode, loss=loss_op, train_op=train_op, training_hooks=[hook],)
def get_vocab(f_path):
word2idx = {}
with open(f_path) as f:
for i, line in enumerate(f):
line = line.rstrip()
word2idx[line] = i
return word2idx
params = {
'model_dir': '../model/lstm_transformer',
'log_path': '../log/lstm_transformer.txt',
'train_path': '../data/train.tsv',
'test_path': '../data/test.tsv',
'vocab_src_path': '../vocab/source.txt',
'vocab_tgt_path': '../vocab/target.txt',
'model_path': '../model/',
'dropout_rate': 0.2,
'rnn_units': 300,
'beam_width': 5,
'activation': tf.nn.relu,
'lr': 4e-4,
'clip_norm': 5.,
'max_dec_len': 80,
'length_penalty': .6,
'buffer_size': 31279,
'batch_size': 32,
'num_patience': 5,
}
class config_model:
hidden_dim = 300
position_embedder_hparams = {
'dim': hidden_dim
}
decoder = {
'dim': hidden_dim,
'num_blocks': 2,
'multihead_attention': {
'num_heads': 8,
'output_dim': hidden_dim,
'use_bias': True,
},
'initializer': {
'type': 'variance_scaling_initializer',
'kwargs': {
'scale': 1.0,
'mode': 'fan_avg',
'distribution': 'uniform',
},
},
'poswise_feedforward': tx.modules.default_transformer_poswise_net_hparams(
output_dim=hidden_dim),
'output_layer_bias': True,
}
params['tgt2idx'] = get_vocab(params['vocab_tgt_path'])
params['idx2tgt'] = {idx: tgt for tgt, idx in params['tgt2idx'].items()}
def is_descending(history: list) -> bool:
history = history[-(params['num_patience']+1):]
for i in range(1, len(history)):
if history[i-1] <= history[i]:
return False
return True
def minimal_test(estimator):
test_str = [['what', 'times', 'are', 'the', 'nutcracker', 'show', 'playing', 'near', 'me']]
predicted = list(estimator.predict(tf.estimator.inputs.numpy_input_fn(
x = np.array(test_str), shuffle = False)))
print('-'*12)
print('minimal test')
print('utterance:', ' '.join(test_str[0]))
predicted = ' '.join([params['idx2tgt'].get(idx, len(params['idx2tgt'])) for idx in predicted[0]])
predicted = predicted.replace('<end>', '').replace('<pad>', '').strip()
print('parsed:', predicted)
print()
try:
nltk.tree.Tree.fromstring(predicted.replace('[ ', '(').replace(' ]', ')')).pretty_print()
except:
pass
print('-'*12)
# Create directory if not exist
Path(os.path.dirname(params['log_path'])).mkdir(exist_ok=True)
Path(params['model_dir']).mkdir(exist_ok=True, parents=True)
# Logging
logger = logging.getLogger('tensorflow')
logger.propagate = False
logger.setLevel(logging.INFO)
fh = logging.FileHandler(params['log_path'])
logger.addHandler(fh)
# Create an estimator
eval_steps = 10 * (params['buffer_size']//params['batch_size'] + 1)
estimator = tf.estimator.Estimator(
model_fn=model_fn,
model_dir=params['model_dir'],
config=tf.estimator.RunConfig(save_checkpoints_steps=eval_steps),
params=params)
best_acc = .0
history_acc = []
tf.enable_eager_execution()
while True:
estimator.train(input_fn=lambda: dataset(is_training=True, params=params))
minimal_test(estimator)
# Evaluation
labels = [label for _, (_, label) in dataset(is_training=False, params=params)]
seq_lens = [tf.argmax(tf.cast(tf.equal(l, 2), tf.int32), axis=1) for l in labels]
labels = [j for i in labels for j in i.numpy()]
seq_lens = [j for i in seq_lens for j in i.numpy()]
preds = list(estimator.predict(input_fn=lambda: dataset(is_training=False, params=params)))
assert len(preds) == len(labels)
res = [np.all(p[:seq_len+1] == l[:seq_len+1]) for p, l, seq_len in zip(preds, labels, seq_lens)]
acc = np.asarray(res).mean()
logger.info("Evaluation: Testing (Exact Match) Accuracy: {:.3f}".format(acc))
history_acc.append(acc)
if acc > best_acc:
best_acc = acc
logger.info("Best (Exact Match) Accuracy: {:.3f}".format(best_acc))
if len(history_acc) > params['num_patience'] and is_descending(history_acc):
logger.info("Testing (Exact Match) Accuracy not improved over {} epochs, Early Stop".format(params['num_patience']))
break
INFO:tensorflow:Using config: {'_model_dir': '../model/lstm_transformer', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': 9780, '_save_checkpoints_secs': None, '_session_config': allow_soft_placement: true
graph_options {
rewrite_options {
meta_optimizer_iterations: ONE
}
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_service': None, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x7fb3bb75b710>, '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/training_util.py:236: Variable.initialized_value (from tensorflow.python.ops.variables) is deprecated and will be removed in a future version.
Instructions for updating:
Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts.
INFO:tensorflow:Calling model_fn.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/util/deprecation.py:507: calling count_nonzero (from tensorflow.python.ops.math_ops) with axis is deprecated and will be removed in a future version.
Instructions for updating:
reduction_indices is deprecated, use axis instead
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
WARNING:tensorflow:From <ipython-input-6-34bae373cd3f>:14: dropout (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.
Instructions for updating:
Use keras.layers.dropout instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/layers/core.py:271: Layer.apply (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.
Instructions for updating:
Please use `layer.__call__` method instead.
WARNING:tensorflow:From <ipython-input-6-34bae373cd3f>:15: dense (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.
Instructions for updating:
Use keras.layers.Dense instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/contrib/rnn/python/ops/lstm_ops.py:597: Layer.add_variable (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.
Instructions for updating:
Please use `layer.add_weight` method instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/module_base.py:72: The name tf.make_template is deprecated. Please use tf.compat.v1.make_template instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/modules/embedders/position_embedders.py:345: The name tf.mod is deprecated. Please use tf.math.mod instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/modules/decoders/transformer_decoders.py:103: The name tf.get_variable_scope is deprecated. Please use tf.compat.v1.get_variable_scope instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/modules/decoders/rnn_decoder_base.py:57: The name tf.get_variable is deprecated. Please use tf.compat.v1.get_variable instead.
WARNING:tensorflow:From /usr/lib/python3.6/pydoc.py:1595: The name tf.layers.Dense is deprecated. Please use tf.compat.v1.layers.Dense instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/core/layers.py:600: The name tf.layers.Layer is deprecated. Please use tf.compat.v1.layers.Layer instead.
WARNING:tensorflow:From /usr/lib/python3.6/pydoc.py:1595: The name tf.layers.Dropout is deprecated. Please use tf.compat.v1.layers.Dropout instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/utils/transformer_attentions.py:105: The name tf.matrix_band_part is deprecated. Please use tf.linalg.band_part instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/modules/decoders/transformer_decoders.py:612: The name tf.AUTO_REUSE is deprecated. Please use tf.compat.v1.AUTO_REUSE instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/module_base.py:129: The name tf.get_collection is deprecated. Please use tf.compat.v1.get_collection instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/module_base.py:130: The name tf.GraphKeys is deprecated. Please use tf.compat.v1.GraphKeys instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/context.py:52: The name tf.get_collection_ref is deprecated. Please use tf.compat.v1.get_collection_ref instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/context.py:55: The name tf.placeholder_with_default is deprecated. Please use tf.compat.v1.placeholder_with_default instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/texar/tf/modules/networks/network_base.py:123: The name tf.layers.BatchNormalization is deprecated. Please use tf.compat.v1.layers.BatchNormalization instead.
WARNING:tensorflow:From <ipython-input-8-46a022234050>:10: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.cast` instead.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/losses/losses_impl.py:121: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.where in 2.0, which has the same broadcast rule as np.where
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 0 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 59.033955, step = 0
INFO:tensorflow:lr = 0.0004
INFO:tensorflow:global_step/sec: 9.45219
INFO:tensorflow:loss = 5.0954924, step = 100 (10.586 sec)
INFO:tensorflow:lr = 0.00039959815 (10.582 sec)
INFO:tensorflow:global_step/sec: 11.9891
INFO:tensorflow:loss = 4.4737344, step = 200 (8.342 sec)
INFO:tensorflow:lr = 0.00039919675 (8.342 sec)
INFO:tensorflow:global_step/sec: 12.0835
INFO:tensorflow:loss = 4.20246, step = 300 (8.271 sec)
INFO:tensorflow:lr = 0.00039879576 (8.272 sec)
INFO:tensorflow:global_step/sec: 12.1546
INFO:tensorflow:loss = 4.0485344, step = 400 (8.226 sec)
INFO:tensorflow:lr = 0.00039839515 (8.226 sec)
INFO:tensorflow:global_step/sec: 12.4508
INFO:tensorflow:loss = 3.9781508, step = 500 (8.034 sec)
INFO:tensorflow:lr = 0.00039799497 (8.033 sec)
INFO:tensorflow:global_step/sec: 12.4297
INFO:tensorflow:loss = 3.8139787, step = 600 (8.042 sec)
INFO:tensorflow:lr = 0.00039759517 (8.042 sec)
INFO:tensorflow:global_step/sec: 12.1124
INFO:tensorflow:loss = 3.5685873, step = 700 (8.262 sec)
INFO:tensorflow:lr = 0.00039719575 (8.262 sec)
INFO:tensorflow:global_step/sec: 12.1605
INFO:tensorflow:loss = 3.525229, step = 800 (8.222 sec)
INFO:tensorflow:lr = 0.0003967968 (8.221 sec)
INFO:tensorflow:global_step/sec: 12.0023
INFO:tensorflow:loss = 3.318803, step = 900 (8.334 sec)
INFO:tensorflow:lr = 0.0003963982 (8.334 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.02963
INFO:tensorflow:loss = 3.2519212, step = 1000 (12.453 sec)
INFO:tensorflow:lr = 0.00039599996 (12.454 sec)
INFO:tensorflow:global_step/sec: 12.1495
INFO:tensorflow:loss = 3.045715, step = 1100 (8.232 sec)
INFO:tensorflow:lr = 0.0003956022 (8.231 sec)
INFO:tensorflow:global_step/sec: 12.0573
INFO:tensorflow:loss = 3.4412713, step = 1200 (8.294 sec)
INFO:tensorflow:lr = 0.00039520478 (8.294 sec)
INFO:tensorflow:global_step/sec: 12.0834
INFO:tensorflow:loss = 2.9590592, step = 1300 (8.273 sec)
INFO:tensorflow:lr = 0.0003948078 (8.275 sec)
INFO:tensorflow:global_step/sec: 12.2286
INFO:tensorflow:loss = 2.948428, step = 1400 (8.179 sec)
INFO:tensorflow:lr = 0.0003944112 (8.178 sec)
INFO:tensorflow:global_step/sec: 12.4073
INFO:tensorflow:loss = 2.8570395, step = 1500 (8.061 sec)
INFO:tensorflow:lr = 0.00039401502 (8.059 sec)
INFO:tensorflow:global_step/sec: 12.1437
INFO:tensorflow:loss = 2.8721683, step = 1600 (8.229 sec)
INFO:tensorflow:lr = 0.0003936192 (8.230 sec)
INFO:tensorflow:global_step/sec: 12.4888
INFO:tensorflow:loss = 2.6866806, step = 1700 (8.007 sec)
INFO:tensorflow:lr = 0.00039322383 (8.006 sec)
INFO:tensorflow:global_step/sec: 12.2946
INFO:tensorflow:loss = 2.5551322, step = 1800 (8.134 sec)
INFO:tensorflow:lr = 0.0003928288 (8.134 sec)
INFO:tensorflow:global_step/sec: 12.5544
INFO:tensorflow:loss = 2.6697116, step = 1900 (7.965 sec)
INFO:tensorflow:lr = 0.0003924342 (7.966 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.853
INFO:tensorflow:loss = 2.5178728, step = 2000 (12.740 sec)
INFO:tensorflow:lr = 0.00039203998 (12.739 sec)
INFO:tensorflow:global_step/sec: 12.1408
INFO:tensorflow:loss = 2.506233, step = 2100 (8.235 sec)
INFO:tensorflow:lr = 0.00039164617 (8.235 sec)
INFO:tensorflow:global_step/sec: 12.46
INFO:tensorflow:loss = 2.2534523, step = 2200 (8.026 sec)
INFO:tensorflow:lr = 0.00039125275 (8.027 sec)
INFO:tensorflow:global_step/sec: 12.586
INFO:tensorflow:loss = 2.0285623, step = 2300 (7.941 sec)
INFO:tensorflow:lr = 0.00039085973 (7.940 sec)
INFO:tensorflow:global_step/sec: 12.5412
INFO:tensorflow:loss = 2.0797236, step = 2400 (7.973 sec)
INFO:tensorflow:lr = 0.00039046712 (7.973 sec)
INFO:tensorflow:global_step/sec: 12.8519
INFO:tensorflow:loss = 2.145553, step = 2500 (7.787 sec)
INFO:tensorflow:lr = 0.00039007486 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.394
INFO:tensorflow:loss = 1.9565487, step = 2600 (8.063 sec)
INFO:tensorflow:lr = 0.00038968303 (8.064 sec)
INFO:tensorflow:global_step/sec: 12.363
INFO:tensorflow:loss = 2.0215652, step = 2700 (8.096 sec)
INFO:tensorflow:lr = 0.00038929158 (8.095 sec)
INFO:tensorflow:global_step/sec: 12.4914
INFO:tensorflow:loss = 2.0743313, step = 2800 (7.998 sec)
INFO:tensorflow:lr = 0.00038890052 (8.002 sec)
INFO:tensorflow:global_step/sec: 12.6555
INFO:tensorflow:loss = 1.9481674, step = 2900 (7.907 sec)
INFO:tensorflow:lr = 0.00038850983 (7.903 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.19896
INFO:tensorflow:loss = 1.8079375, step = 3000 (12.195 sec)
INFO:tensorflow:lr = 0.0003881196 (12.195 sec)
INFO:tensorflow:global_step/sec: 12.647
INFO:tensorflow:loss = 1.858728, step = 3100 (7.910 sec)
INFO:tensorflow:lr = 0.0003877297 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.7231
INFO:tensorflow:loss = 1.7418987, step = 3200 (7.853 sec)
INFO:tensorflow:lr = 0.00038734023 (7.853 sec)
INFO:tensorflow:global_step/sec: 12.633
INFO:tensorflow:loss = 1.8075138, step = 3300 (7.915 sec)
INFO:tensorflow:lr = 0.00038695114 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.6876
INFO:tensorflow:loss = 1.9535941, step = 3400 (7.886 sec)
INFO:tensorflow:lr = 0.00038656246 (7.886 sec)
INFO:tensorflow:global_step/sec: 12.8031
INFO:tensorflow:loss = 1.7075312, step = 3500 (7.807 sec)
INFO:tensorflow:lr = 0.00038617416 (7.808 sec)
INFO:tensorflow:global_step/sec: 12.5066
INFO:tensorflow:loss = 1.8058743, step = 3600 (7.997 sec)
INFO:tensorflow:lr = 0.0003857862 (7.996 sec)
INFO:tensorflow:global_step/sec: 12.8845
INFO:tensorflow:loss = 1.6459893, step = 3700 (7.764 sec)
INFO:tensorflow:lr = 0.00038539866 (7.764 sec)
INFO:tensorflow:global_step/sec: 12.6028
INFO:tensorflow:loss = 1.8697182, step = 3800 (7.930 sec)
INFO:tensorflow:lr = 0.00038501152 (7.930 sec)
INFO:tensorflow:global_step/sec: 12.6564
INFO:tensorflow:loss = 1.7144407, step = 3900 (7.904 sec)
INFO:tensorflow:lr = 0.00038462476 (7.905 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.39316
INFO:tensorflow:loss = 1.6077055, step = 4000 (11.917 sec)
INFO:tensorflow:lr = 0.0003842384 (11.915 sec)
INFO:tensorflow:global_step/sec: 12.8114
INFO:tensorflow:loss = 1.6767513, step = 4100 (7.804 sec)
INFO:tensorflow:lr = 0.00038385243 (7.808 sec)
INFO:tensorflow:global_step/sec: 12.63
INFO:tensorflow:loss = 1.6651752, step = 4200 (7.919 sec)
INFO:tensorflow:lr = 0.00038346683 (7.915 sec)
INFO:tensorflow:global_step/sec: 12.8231
INFO:tensorflow:loss = 1.574295, step = 4300 (7.798 sec)
INFO:tensorflow:lr = 0.00038308164 (7.798 sec)
INFO:tensorflow:global_step/sec: 12.4846
INFO:tensorflow:loss = 1.6911457, step = 4400 (8.006 sec)
INFO:tensorflow:lr = 0.00038269683 (8.009 sec)
INFO:tensorflow:global_step/sec: 12.6322
INFO:tensorflow:loss = 1.6363447, step = 4500 (7.920 sec)
INFO:tensorflow:lr = 0.0003823124 (7.917 sec)
INFO:tensorflow:global_step/sec: 12.8063
INFO:tensorflow:loss = 1.7158389, step = 4600 (7.804 sec)
INFO:tensorflow:lr = 0.00038192834 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.6763
INFO:tensorflow:loss = 1.6156652, step = 4700 (7.889 sec)
INFO:tensorflow:lr = 0.00038154467 (7.890 sec)
INFO:tensorflow:global_step/sec: 12.1828
INFO:tensorflow:loss = 1.4845752, step = 4800 (8.208 sec)
INFO:tensorflow:lr = 0.0003811614 (8.207 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.4912
INFO:tensorflow:loss = 1.6173213, step = 4900 (11.777 sec)
INFO:tensorflow:lr = 0.0003807785 (11.777 sec)
INFO:tensorflow:global_step/sec: 12.9355
INFO:tensorflow:loss = 1.5071679, step = 5000 (7.735 sec)
INFO:tensorflow:lr = 0.00038039603 (7.736 sec)
INFO:tensorflow:global_step/sec: 12.8336
INFO:tensorflow:loss = 1.5181412, step = 5100 (7.788 sec)
INFO:tensorflow:lr = 0.0003800139 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.678
INFO:tensorflow:loss = 1.6296129, step = 5200 (7.890 sec)
INFO:tensorflow:lr = 0.00037963217 (7.893 sec)
INFO:tensorflow:global_step/sec: 12.4484
INFO:tensorflow:loss = 1.5062387, step = 5300 (8.033 sec)
INFO:tensorflow:lr = 0.00037925082 (8.031 sec)
INFO:tensorflow:global_step/sec: 12.8709
INFO:tensorflow:loss = 1.4886248, step = 5400 (7.771 sec)
INFO:tensorflow:lr = 0.00037886985 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.6909
INFO:tensorflow:loss = 1.4882165, step = 5500 (7.874 sec)
INFO:tensorflow:lr = 0.00037848923 (7.877 sec)
INFO:tensorflow:global_step/sec: 12.6324
INFO:tensorflow:loss = 1.48171, step = 5600 (7.917 sec)
INFO:tensorflow:lr = 0.00037810905 (7.913 sec)
INFO:tensorflow:global_step/sec: 12.8959
INFO:tensorflow:loss = 1.5500232, step = 5700 (7.757 sec)
INFO:tensorflow:lr = 0.00037772924 (7.757 sec)
INFO:tensorflow:global_step/sec: 12.8081
INFO:tensorflow:loss = 1.4949408, step = 5800 (7.809 sec)
INFO:tensorflow:lr = 0.0003773498 (7.808 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.87373
INFO:tensorflow:loss = 1.5293541, step = 5900 (11.270 sec)
INFO:tensorflow:lr = 0.00037697077 (11.272 sec)
INFO:tensorflow:global_step/sec: 12.9509
INFO:tensorflow:loss = 1.5244159, step = 6000 (7.716 sec)
INFO:tensorflow:lr = 0.00037659207 (7.717 sec)
INFO:tensorflow:global_step/sec: 12.8297
INFO:tensorflow:loss = 1.4912233, step = 6100 (7.795 sec)
INFO:tensorflow:lr = 0.00037621378 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.8172
INFO:tensorflow:loss = 1.4585661, step = 6200 (7.803 sec)
INFO:tensorflow:lr = 0.00037583587 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.8078
INFO:tensorflow:loss = 1.4709327, step = 6300 (7.807 sec)
INFO:tensorflow:lr = 0.00037545833 (7.810 sec)
INFO:tensorflow:global_step/sec: 12.768
INFO:tensorflow:loss = 1.5252771, step = 6400 (7.836 sec)
INFO:tensorflow:lr = 0.00037508114 (7.834 sec)
INFO:tensorflow:global_step/sec: 12.4801
INFO:tensorflow:loss = 1.4626974, step = 6500 (8.016 sec)
INFO:tensorflow:lr = 0.00037470434 (8.015 sec)
INFO:tensorflow:global_step/sec: 12.4797
INFO:tensorflow:loss = 1.4680932, step = 6600 (8.011 sec)
INFO:tensorflow:lr = 0.00037432797 (8.011 sec)
INFO:tensorflow:global_step/sec: 12.6182
INFO:tensorflow:loss = 1.4867874, step = 6700 (7.920 sec)
INFO:tensorflow:lr = 0.00037395195 (7.921 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.7869
INFO:tensorflow:loss = 1.5120995, step = 6800 (8.494 sec)
INFO:tensorflow:lr = 0.0003735763 (8.497 sec)
INFO:tensorflow:global_step/sec: 10.8646
INFO:tensorflow:loss = 1.5093869, step = 6900 (9.195 sec)
INFO:tensorflow:lr = 0.00037320104 (9.192 sec)
INFO:tensorflow:global_step/sec: 12.6021
INFO:tensorflow:loss = 1.4627079, step = 7000 (7.934 sec)
INFO:tensorflow:lr = 0.00037282615 (7.938 sec)
INFO:tensorflow:global_step/sec: 12.4589
INFO:tensorflow:loss = 1.4051437, step = 7100 (8.026 sec)
INFO:tensorflow:lr = 0.00037245164 (8.022 sec)
INFO:tensorflow:global_step/sec: 12.9606
INFO:tensorflow:loss = 1.4104779, step = 7200 (7.716 sec)
INFO:tensorflow:lr = 0.0003720775 (7.716 sec)
INFO:tensorflow:global_step/sec: 12.5317
INFO:tensorflow:loss = 1.4525088, step = 7300 (7.979 sec)
INFO:tensorflow:lr = 0.00037170373 (7.980 sec)
INFO:tensorflow:global_step/sec: 12.8884
INFO:tensorflow:loss = 1.4862131, step = 7400 (7.763 sec)
INFO:tensorflow:lr = 0.00037133033 (7.763 sec)
INFO:tensorflow:global_step/sec: 12.3808
INFO:tensorflow:loss = 1.4874706, step = 7500 (8.079 sec)
INFO:tensorflow:lr = 0.0003709573 (8.078 sec)
INFO:tensorflow:global_step/sec: 12.6027
INFO:tensorflow:loss = 1.4101737, step = 7600 (7.931 sec)
INFO:tensorflow:lr = 0.0003705847 (7.935 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.5763
INFO:tensorflow:loss = 1.4103632, step = 7700 (7.951 sec)
INFO:tensorflow:lr = 0.00037021242 (7.948 sec)
INFO:tensorflow:global_step/sec: 11.0486
INFO:tensorflow:loss = 1.4301885, step = 7800 (9.050 sec)
INFO:tensorflow:lr = 0.00036984053 (9.049 sec)
INFO:tensorflow:global_step/sec: 12.9262
INFO:tensorflow:loss = 1.4219362, step = 7900 (7.738 sec)
INFO:tensorflow:lr = 0.00036946902 (7.743 sec)
INFO:tensorflow:global_step/sec: 12.6925
INFO:tensorflow:loss = 1.4855167, step = 8000 (7.877 sec)
INFO:tensorflow:lr = 0.0003690979 (7.872 sec)
INFO:tensorflow:global_step/sec: 12.5745
INFO:tensorflow:loss = 1.3784947, step = 8100 (7.956 sec)
INFO:tensorflow:lr = 0.00036872714 (7.956 sec)
INFO:tensorflow:global_step/sec: 12.545
INFO:tensorflow:loss = 1.4119531, step = 8200 (7.970 sec)
INFO:tensorflow:lr = 0.0003683567 (7.971 sec)
INFO:tensorflow:global_step/sec: 12.5777
INFO:tensorflow:loss = 1.4218488, step = 8300 (7.949 sec)
INFO:tensorflow:lr = 0.00036798668 (7.950 sec)
INFO:tensorflow:global_step/sec: 12.7638
INFO:tensorflow:loss = 1.4013296, step = 8400 (7.835 sec)
INFO:tensorflow:lr = 0.00036761703 (7.833 sec)
INFO:tensorflow:global_step/sec: 12.6604
INFO:tensorflow:loss = 1.3762985, step = 8500 (7.900 sec)
INFO:tensorflow:lr = 0.00036724776 (7.900 sec)
INFO:tensorflow:global_step/sec: 13.0189
INFO:tensorflow:loss = 1.4109349, step = 8600 (7.682 sec)
INFO:tensorflow:lr = 0.00036687887 (7.683 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.6167
INFO:tensorflow:loss = 1.4246565, step = 8700 (8.613 sec)
INFO:tensorflow:lr = 0.0003665103 (8.619 sec)
INFO:tensorflow:global_step/sec: 11.3619
INFO:tensorflow:loss = 1.3846593, step = 8800 (8.798 sec)
INFO:tensorflow:lr = 0.00036614214 (8.791 sec)
INFO:tensorflow:global_step/sec: 12.544
INFO:tensorflow:loss = 1.4037898, step = 8900 (7.972 sec)
INFO:tensorflow:lr = 0.00036577435 (7.972 sec)
INFO:tensorflow:global_step/sec: 12.6411
INFO:tensorflow:loss = 1.4317508, step = 9000 (7.911 sec)
INFO:tensorflow:lr = 0.00036540694 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.762
INFO:tensorflow:loss = 1.4514229, step = 9100 (7.842 sec)
INFO:tensorflow:lr = 0.00036503986 (7.842 sec)
INFO:tensorflow:global_step/sec: 12.8531
INFO:tensorflow:loss = 1.3912779, step = 9200 (7.773 sec)
INFO:tensorflow:lr = 0.00036467315 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.6416
INFO:tensorflow:loss = 1.4078438, step = 9300 (7.910 sec)
INFO:tensorflow:lr = 0.00036430682 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.7305
INFO:tensorflow:loss = 1.3546025, step = 9400 (7.854 sec)
INFO:tensorflow:lr = 0.00036394087 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.6416
INFO:tensorflow:loss = 1.376714, step = 9500 (7.913 sec)
INFO:tensorflow:lr = 0.0003635753 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.5347
INFO:tensorflow:loss = 1.3942187, step = 9600 (7.973 sec)
INFO:tensorflow:lr = 0.00036321007 (7.973 sec)
INFO:tensorflow:global_step/sec: 13.0655
INFO:tensorflow:loss = 1.3802559, step = 9700 (7.654 sec)
INFO:tensorflow:lr = 0.00036284523 (7.654 sec)
INFO:tensorflow:Saving checkpoints for 9780 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.3757502.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_estimator/python/estimator/inputs/queues/feeding_queue_runner.py:62: QueueRunner.__init__ (from tensorflow.python.training.queue_runner_impl) is deprecated and will be removed in a future version.
Instructions for updating:
To construct input pipelines, use the `tf.data` module.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_estimator/python/estimator/inputs/queues/feeding_functions.py:500: add_queue_runner (from tensorflow.python.training.queue_runner_impl) is deprecated and will be removed in a future version.
Instructions for updating:
To construct input pipelines, use the `tf.data` module.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-9780
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py:882: start_queue_runners (from tensorflow.python.training.queue_runner_impl) is deprecated and will be removed in a future version.
Instructions for updating:
To construct input pipelines, use the `tf.data` module.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-9780
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.614
INFO:tensorflow:Best (Exact Match) Accuracy: 0.614
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:tensorflow:Saver not created because there are no variables in the graph to restore [<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>, <tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>, <tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>, <tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>, <tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>, <tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>, <tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>] INFO:tensorflow:Done calling model_fn. INFO:tensorflow:Create CheckpointSaverHook. INFO:tensorflow:Graph was finalized. INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-9780 WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/saver.py:1069: get_checkpoint_mtimes (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version. Instructions for updating: Use standard file utilities to get mtimes. INFO:tensorflow:Running local_init_op. INFO:tensorflow:Done running local_init_op. INFO:tensorflow:Saving checkpoints for 9780 into ../model/lstm_transformer/model.ckpt. Reading ../data/train.tsv INFO:tensorflow:loss = 1.3950752, step = 9780 INFO:tensorflow:lr = 0.00036255358 INFO:tensorflow:global_step/sec: 9.89531 INFO:tensorflow:loss = 1.38784, step = 9880 (10.114 sec) INFO:tensorflow:lr = 0.0003621894 (10.114 sec) INFO:tensorflow:global_step/sec: 12.511 INFO:tensorflow:loss = 1.3594651, step = 9980 (7.991 sec) INFO:tensorflow:lr = 0.00036182557 (7.986 sec) INFO:tensorflow:global_step/sec: 12.6818 INFO:tensorflow:loss = 1.3496534, step = 10080 (7.885 sec) INFO:tensorflow:lr = 0.0003614621 (7.885 sec) INFO:tensorflow:global_step/sec: 12.8195 INFO:tensorflow:loss = 1.3859842, step = 10180 (7.800 sec) INFO:tensorflow:lr = 0.00036109902 (7.801 sec) INFO:tensorflow:global_step/sec: 12.7904 INFO:tensorflow:loss = 1.4064968, step = 10280 (7.821 sec) INFO:tensorflow:lr = 0.0003607363 (7.820 sec) INFO:tensorflow:global_step/sec: 12.9878 INFO:tensorflow:loss = 1.3514677, step = 10380 (7.697 sec) INFO:tensorflow:lr = 0.00036037393 (7.697 sec) INFO:tensorflow:global_step/sec: 12.7001 INFO:tensorflow:loss = 1.3718523, step = 10480 (7.877 sec) INFO:tensorflow:lr = 0.0003600119 (7.879 sec) INFO:tensorflow:global_step/sec: 12.4932 INFO:tensorflow:loss = 1.3550223, step = 10580 (8.002 sec) INFO:tensorflow:lr = 0.00035965024 (8.002 sec) INFO:tensorflow:global_step/sec: 12.7582 INFO:tensorflow:loss = 1.3919998, step = 10680 (7.836 sec) INFO:tensorflow:lr = 0.00035928897 (7.834 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 8.20005 INFO:tensorflow:loss = 1.3931295, step = 10780 (12.198 sec) INFO:tensorflow:lr = 0.00035892805 (12.199 sec) INFO:tensorflow:global_step/sec: 12.7685 INFO:tensorflow:loss = 1.3510262, step = 10880 (7.829 sec) INFO:tensorflow:lr = 0.00035856751 (7.828 sec) INFO:tensorflow:global_step/sec: 12.7663 INFO:tensorflow:loss = 1.398536, step = 10980 (7.836 sec) INFO:tensorflow:lr = 0.00035820733 (7.836 sec) INFO:tensorflow:global_step/sec: 12.6607 INFO:tensorflow:loss = 1.3688757, step = 11080 (7.898 sec) INFO:tensorflow:lr = 0.0003578475 (7.898 sec) INFO:tensorflow:global_step/sec: 12.846 INFO:tensorflow:loss = 1.3669326, step = 11180 (7.786 sec) INFO:tensorflow:lr = 0.00035748802 (7.785 sec) INFO:tensorflow:global_step/sec: 13.0049 INFO:tensorflow:loss = 1.3798718, step = 11280 (7.688 sec) INFO:tensorflow:lr = 0.0003571289 (7.689 sec) INFO:tensorflow:global_step/sec: 12.5146 INFO:tensorflow:loss = 1.361347, step = 11380 (7.987 sec) INFO:tensorflow:lr = 0.0003567702 (7.988 sec) INFO:tensorflow:global_step/sec: 12.5729 INFO:tensorflow:loss = 1.353298, step = 11480 (7.958 sec) INFO:tensorflow:lr = 0.0003564118 (7.957 sec) INFO:tensorflow:global_step/sec: 12.7091 INFO:tensorflow:loss = 1.3753021, step = 11580 (7.863 sec) INFO:tensorflow:lr = 0.00035605376 (7.864 sec) INFO:tensorflow:global_step/sec: 12.7804 INFO:tensorflow:loss = 1.3804079, step = 11680 (7.830 sec) INFO:tensorflow:lr = 0.0003556961 (7.829 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 8.14235 INFO:tensorflow:loss = 1.3768957, step = 11780 (12.278 sec) INFO:tensorflow:lr = 0.00035533877 (12.278 sec) INFO:tensorflow:global_step/sec: 12.6599 INFO:tensorflow:loss = 1.3538108, step = 11880 (7.902 sec) INFO:tensorflow:lr = 0.0003549818 (7.902 sec) INFO:tensorflow:global_step/sec: 12.5448 INFO:tensorflow:loss = 1.3600917, step = 11980 (7.973 sec) INFO:tensorflow:lr = 0.00035462523 (7.975 sec) INFO:tensorflow:global_step/sec: 12.8167 INFO:tensorflow:loss = 1.3522173, step = 12080 (7.801 sec) INFO:tensorflow:lr = 0.00035426902 (7.805 sec) INFO:tensorflow:global_step/sec: 12.7571 INFO:tensorflow:loss = 1.3719348, step = 12180 (7.838 sec) INFO:tensorflow:lr = 0.00035391314 (7.832 sec) INFO:tensorflow:global_step/sec: 12.9242 INFO:tensorflow:loss = 1.3397022, step = 12280 (7.734 sec) INFO:tensorflow:lr = 0.00035355764 (7.735 sec) INFO:tensorflow:global_step/sec: 12.7453 INFO:tensorflow:loss = 1.3814683, step = 12380 (7.848 sec) INFO:tensorflow:lr = 0.0003532025 (7.847 sec) INFO:tensorflow:global_step/sec: 12.3341 INFO:tensorflow:loss = 1.3348244, step = 12480 (8.105 sec) INFO:tensorflow:lr = 0.00035284765 (8.105 sec) INFO:tensorflow:global_step/sec: 12.9432 INFO:tensorflow:loss = 1.3358412, step = 12580 (7.731 sec) INFO:tensorflow:lr = 0.0003524932 (7.731 sec) INFO:tensorflow:global_step/sec: 12.7795 INFO:tensorflow:loss = 1.3471819, step = 12680 (7.821 sec) INFO:tensorflow:lr = 0.00035213912 (7.822 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 8.29116 INFO:tensorflow:loss = 1.3598224, step = 12780 (12.059 sec) INFO:tensorflow:lr = 0.0003517854 (12.058 sec) INFO:tensorflow:global_step/sec: 12.5836 INFO:tensorflow:loss = 1.3555378, step = 12880 (7.951 sec) INFO:tensorflow:lr = 0.00035143204 (7.951 sec) INFO:tensorflow:global_step/sec: 12.7837 INFO:tensorflow:loss = 1.3521165, step = 12980 (7.819 sec) INFO:tensorflow:lr = 0.000351079 (7.819 sec) INFO:tensorflow:global_step/sec: 12.7323 INFO:tensorflow:loss = 1.3695495, step = 13080 (7.859 sec) INFO:tensorflow:lr = 0.00035072633 (7.861 sec) INFO:tensorflow:global_step/sec: 13.0184 INFO:tensorflow:loss = 1.3140963, step = 13180 (7.676 sec) INFO:tensorflow:lr = 0.00035037403 (7.675 sec) INFO:tensorflow:global_step/sec: 12.6429 INFO:tensorflow:loss = 1.3322172, step = 13280 (7.910 sec) INFO:tensorflow:lr = 0.00035002205 (7.913 sec) INFO:tensorflow:global_step/sec: 12.6645 INFO:tensorflow:loss = 1.3830324, step = 13380 (7.901 sec) INFO:tensorflow:lr = 0.00034967044 (7.897 sec) INFO:tensorflow:global_step/sec: 12.349 INFO:tensorflow:loss = 1.3274176, step = 13480 (8.093 sec) INFO:tensorflow:lr = 0.0003493192 (8.099 sec) INFO:tensorflow:global_step/sec: 12.8935 INFO:tensorflow:loss = 1.3599035, step = 13580 (7.758 sec) INFO:tensorflow:lr = 0.00034896831 (7.753 sec) INFO:tensorflow:global_step/sec: 12.5766 INFO:tensorflow:loss = 1.3469765, step = 13680 (7.953 sec) INFO:tensorflow:lr = 0.00034861776 (7.953 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 8.4964 INFO:tensorflow:loss = 1.3144921, step = 13780 (11.765 sec) INFO:tensorflow:lr = 0.00034826755 (11.764 sec) INFO:tensorflow:global_step/sec: 12.4808 INFO:tensorflow:loss = 1.4038187, step = 13880 (8.014 sec) INFO:tensorflow:lr = 0.00034791772 (8.016 sec) INFO:tensorflow:global_step/sec: 12.8326 INFO:tensorflow:loss = 1.3301687, step = 13980 (7.792 sec) INFO:tensorflow:lr = 0.00034756822 (7.791 sec) INFO:tensorflow:global_step/sec: 12.692 INFO:tensorflow:loss = 1.3259747, step = 14080 (7.882 sec) INFO:tensorflow:lr = 0.0003472191 (7.882 sec) INFO:tensorflow:global_step/sec: 12.7396 INFO:tensorflow:loss = 1.3204691, step = 14180 (7.846 sec) INFO:tensorflow:lr = 0.0003468703 (7.849 sec) INFO:tensorflow:global_step/sec: 12.5856 INFO:tensorflow:loss = 1.3055412, step = 14280 (7.946 sec) INFO:tensorflow:lr = 0.00034652182 (7.943 sec) INFO:tensorflow:global_step/sec: 12.639 INFO:tensorflow:loss = 1.3440651, step = 14380 (7.917 sec) INFO:tensorflow:lr = 0.00034617377 (7.916 sec) INFO:tensorflow:global_step/sec: 12.5423 INFO:tensorflow:loss = 1.3381974, step = 14480 (7.967 sec) INFO:tensorflow:lr = 0.000345826 (7.969 sec) INFO:tensorflow:global_step/sec: 12.6293 INFO:tensorflow:loss = 1.3422388, step = 14580 (7.921 sec) INFO:tensorflow:lr = 0.00034547862 (7.921 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 8.63617 INFO:tensorflow:loss = 1.3256061, step = 14680 (11.581 sec) INFO:tensorflow:lr = 0.0003451316 (11.581 sec) INFO:tensorflow:global_step/sec: 13.0441 INFO:tensorflow:loss = 1.3087891, step = 14780 (7.665 sec) INFO:tensorflow:lr = 0.00034478487 (7.664 sec) INFO:tensorflow:global_step/sec: 12.7367 INFO:tensorflow:loss = 1.343419, step = 14880 (7.853 sec) INFO:tensorflow:lr = 0.00034443854 (7.853 sec) INFO:tensorflow:global_step/sec: 12.7033 INFO:tensorflow:loss = 1.3251257, step = 14980 (7.869 sec) INFO:tensorflow:lr = 0.00034409255 (7.869 sec) INFO:tensorflow:global_step/sec: 12.6804 INFO:tensorflow:loss = 1.34874, step = 15080 (7.890 sec) INFO:tensorflow:lr = 0.00034374685 (7.889 sec) INFO:tensorflow:global_step/sec: 12.414 INFO:tensorflow:loss = 1.3258908, step = 15180 (8.049 sec) INFO:tensorflow:lr = 0.00034340157 (8.050 sec) INFO:tensorflow:global_step/sec: 13.0191 INFO:tensorflow:loss = 1.331736, step = 15280 (7.681 sec) INFO:tensorflow:lr = 0.00034305663 (7.684 sec) INFO:tensorflow:global_step/sec: 12.7453 INFO:tensorflow:loss = 1.3291036, step = 15380 (7.851 sec) INFO:tensorflow:lr = 0.000342712 (7.847 sec) INFO:tensorflow:global_step/sec: 12.5726 INFO:tensorflow:loss = 1.360546, step = 15480 (7.955 sec) INFO:tensorflow:lr = 0.00034236774 (7.955 sec) INFO:tensorflow:global_step/sec: 12.4596 INFO:tensorflow:loss = 1.3105683, step = 15580 (8.021 sec) INFO:tensorflow:lr = 0.00034202382 (8.021 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 9.10404 INFO:tensorflow:loss = 1.306641, step = 15680 (10.988 sec) INFO:tensorflow:lr = 0.00034168025 (10.990 sec) INFO:tensorflow:global_step/sec: 12.6049 INFO:tensorflow:loss = 1.3161285, step = 15780 (7.932 sec) INFO:tensorflow:lr = 0.00034133703 (7.931 sec) INFO:tensorflow:global_step/sec: 13.1158 INFO:tensorflow:loss = 1.331718, step = 15880 (7.627 sec) INFO:tensorflow:lr = 0.00034099416 (7.627 sec) INFO:tensorflow:global_step/sec: 12.7958 INFO:tensorflow:loss = 1.3309196, step = 15980 (7.811 sec) INFO:tensorflow:lr = 0.00034065163 (7.810 sec) INFO:tensorflow:global_step/sec: 12.8154 INFO:tensorflow:loss = 1.3363672, step = 16080 (7.803 sec) INFO:tensorflow:lr = 0.00034030943 (7.803 sec) INFO:tensorflow:global_step/sec: 12.8214 INFO:tensorflow:loss = 1.3402402, step = 16180 (7.803 sec) INFO:tensorflow:lr = 0.00033996755 (7.803 sec) INFO:tensorflow:global_step/sec: 12.845 INFO:tensorflow:loss = 1.3548362, step = 16280 (7.784 sec) INFO:tensorflow:lr = 0.00033962607 (7.785 sec) INFO:tensorflow:global_step/sec: 12.3373 INFO:tensorflow:loss = 1.3413169, step = 16380 (8.103 sec) INFO:tensorflow:lr = 0.0003392849 (8.102 sec) INFO:tensorflow:global_step/sec: 12.5631 INFO:tensorflow:loss = 1.3316003, step = 16480 (7.964 sec) INFO:tensorflow:lr = 0.00033894408 (7.964 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 12.1431 INFO:tensorflow:loss = 1.3373045, step = 16580 (8.238 sec) INFO:tensorflow:lr = 0.0003386036 (8.244 sec) INFO:tensorflow:global_step/sec: 10.6592 INFO:tensorflow:loss = 1.3188674, step = 16680 (9.380 sec) INFO:tensorflow:lr = 0.00033826346 (9.374 sec) INFO:tensorflow:global_step/sec: 12.777 INFO:tensorflow:loss = 1.2873152, step = 16780 (7.825 sec) INFO:tensorflow:lr = 0.00033792367 (7.825 sec) INFO:tensorflow:global_step/sec: 12.5919 INFO:tensorflow:loss = 1.3186045, step = 16880 (7.937 sec) INFO:tensorflow:lr = 0.00033758424 (7.937 sec) INFO:tensorflow:global_step/sec: 12.728 INFO:tensorflow:loss = 1.3294945, step = 16980 (7.857 sec) INFO:tensorflow:lr = 0.0003372451 (7.857 sec) INFO:tensorflow:global_step/sec: 12.9323 INFO:tensorflow:loss = 1.3417977, step = 17080 (7.738 sec) INFO:tensorflow:lr = 0.00033690632 (7.738 sec) INFO:tensorflow:global_step/sec: 12.7768 INFO:tensorflow:loss = 1.3123543, step = 17180 (7.826 sec) INFO:tensorflow:lr = 0.0003365679 (7.825 sec) INFO:tensorflow:global_step/sec: 12.7613 INFO:tensorflow:loss = 1.319809, step = 17280 (7.834 sec) INFO:tensorflow:lr = 0.0003362298 (7.834 sec) INFO:tensorflow:global_step/sec: 12.5586 INFO:tensorflow:loss = 1.31707, step = 17380 (7.961 sec) INFO:tensorflow:lr = 0.00033589205 (7.961 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 12.5363 INFO:tensorflow:loss = 1.3020943, step = 17480 (7.982 sec) INFO:tensorflow:lr = 0.00033555465 (7.984 sec) INFO:tensorflow:global_step/sec: 10.9413 INFO:tensorflow:loss = 1.3228619, step = 17580 (9.138 sec) INFO:tensorflow:lr = 0.00033521757 (9.137 sec) INFO:tensorflow:global_step/sec: 12.6287 INFO:tensorflow:loss = 1.3076676, step = 17680 (7.919 sec) INFO:tensorflow:lr = 0.00033488084 (7.918 sec) INFO:tensorflow:global_step/sec: 12.7388 INFO:tensorflow:loss = 1.301673, step = 17780 (7.849 sec) INFO:tensorflow:lr = 0.0003345444 (7.850 sec) INFO:tensorflow:global_step/sec: 12.7397 INFO:tensorflow:loss = 1.3108275, step = 17880 (7.846 sec) INFO:tensorflow:lr = 0.00033420837 (7.846 sec) INFO:tensorflow:global_step/sec: 12.7183 INFO:tensorflow:loss = 1.3072705, step = 17980 (7.865 sec) INFO:tensorflow:lr = 0.00033387265 (7.867 sec) INFO:tensorflow:global_step/sec: 12.823 INFO:tensorflow:loss = 1.3254036, step = 18080 (7.801 sec) INFO:tensorflow:lr = 0.00033353726 (7.800 sec) INFO:tensorflow:global_step/sec: 12.1744 INFO:tensorflow:loss = 1.299603, step = 18180 (8.208 sec) INFO:tensorflow:lr = 0.00033320222 (8.208 sec) INFO:tensorflow:global_step/sec: 12.6291 INFO:tensorflow:loss = 1.3302066, step = 18280 (7.918 sec) INFO:tensorflow:lr = 0.0003328675 (7.917 sec) INFO:tensorflow:global_step/sec: 12.6967 INFO:tensorflow:loss = 1.318776, step = 18380 (7.879 sec) INFO:tensorflow:lr = 0.00033253315 (7.881 sec) Reading ../data/train.tsv INFO:tensorflow:global_step/sec: 11.8313 INFO:tensorflow:loss = 1.2908607, step = 18480 (8.451 sec) INFO:tensorflow:lr = 0.00033219912 (8.454 sec) INFO:tensorflow:global_step/sec: 11.3725 INFO:tensorflow:loss = 1.3153081, step = 18580 (8.796 sec) INFO:tensorflow:lr = 0.0003318654 (8.792 sec) INFO:tensorflow:global_step/sec: 12.7336 INFO:tensorflow:loss = 1.3217913, step = 18680 (7.853 sec) INFO:tensorflow:lr = 0.00033153204 (7.852 sec) INFO:tensorflow:global_step/sec: 12.6533 INFO:tensorflow:loss = 1.3215, step = 18780 (7.904 sec) INFO:tensorflow:lr = 0.00033119903 (7.905 sec) INFO:tensorflow:global_step/sec: 12.4503 INFO:tensorflow:loss = 1.3031405, step = 18880 (8.031 sec) INFO:tensorflow:lr = 0.00033086629 (8.031 sec) INFO:tensorflow:global_step/sec: 12.6567 INFO:tensorflow:loss = 1.2904049, step = 18980 (7.896 sec) INFO:tensorflow:lr = 0.00033053392 (7.897 sec) INFO:tensorflow:global_step/sec: 12.6398 INFO:tensorflow:loss = 1.3116907, step = 19080 (7.912 sec) INFO:tensorflow:lr = 0.00033020187 (7.911 sec) INFO:tensorflow:global_step/sec: 12.6908 INFO:tensorflow:loss = 1.3194084, step = 19180 (7.883 sec) INFO:tensorflow:lr = 0.00032987018 (7.883 sec) INFO:tensorflow:global_step/sec: 12.8419 INFO:tensorflow:loss = 1.3328631, step = 19280 (7.784 sec) INFO:tensorflow:lr = 0.00032953883 (7.784 sec) INFO:tensorflow:global_step/sec: 13.1392 INFO:tensorflow:loss = 1.3144226, step = 19380 (7.611 sec) INFO:tensorflow:lr = 0.0003292078 (7.611 sec) INFO:tensorflow:global_step/sec: 12.6009 INFO:tensorflow:loss = 1.3241731, step = 19480 (7.936 sec) INFO:tensorflow:lr = 0.0003288771 (7.936 sec) INFO:tensorflow:Saving checkpoints for 19560 into ../model/lstm_transformer/model.ckpt. INFO:tensorflow:Loss for final step: 1.3418434. INFO:tensorflow:Calling model_fn. INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:tensorflow:Done calling model_fn. INFO:tensorflow:Graph was finalized. INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-19560 INFO:tensorflow:Running local_init_op. INFO:tensorflow:Done running local_init_op. ------------ minimal test utterance: what times are the nutcracker show playing near me parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ------------ Reading ../data/test.tsv INFO:tensorflow:Calling model_fn. INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:tensorflow:Saver not created because there are no variables in the graph to restore INFO:tensorflow:Done calling model_fn. INFO:tensorflow:Graph was finalized. INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-19560 INFO:tensorflow:Running local_init_op. INFO:tensorflow:Done running local_init_op. Reading ../data/test.tsv INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.664 INFO:tensorflow:Best (Exact Match) Accuracy: 0.664 INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-19560
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 19560 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.3048319, step = 19560
INFO:tensorflow:lr = 0.00032861278
INFO:tensorflow:global_step/sec: 10.2833
INFO:tensorflow:loss = 1.312456, step = 19660 (9.729 sec)
INFO:tensorflow:lr = 0.00032828268 (9.725 sec)
INFO:tensorflow:global_step/sec: 12.8592
INFO:tensorflow:loss = 1.3122357, step = 19760 (7.778 sec)
INFO:tensorflow:lr = 0.00032795293 (7.776 sec)
INFO:tensorflow:global_step/sec: 12.5629
INFO:tensorflow:loss = 1.315895, step = 19860 (7.960 sec)
INFO:tensorflow:lr = 0.00032762348 (7.959 sec)
INFO:tensorflow:global_step/sec: 12.982
INFO:tensorflow:loss = 1.2998242, step = 19960 (7.703 sec)
INFO:tensorflow:lr = 0.00032729437 (7.703 sec)
INFO:tensorflow:global_step/sec: 12.8239
INFO:tensorflow:loss = 1.3176674, step = 20060 (7.797 sec)
INFO:tensorflow:lr = 0.00032696562 (7.797 sec)
INFO:tensorflow:global_step/sec: 12.4351
INFO:tensorflow:loss = 1.2875775, step = 20160 (8.039 sec)
INFO:tensorflow:lr = 0.00032663712 (8.042 sec)
INFO:tensorflow:global_step/sec: 12.6837
INFO:tensorflow:loss = 1.2956387, step = 20260 (7.884 sec)
INFO:tensorflow:lr = 0.000326309 (7.881 sec)
INFO:tensorflow:global_step/sec: 12.5531
INFO:tensorflow:loss = 1.3311186, step = 20360 (7.969 sec)
INFO:tensorflow:lr = 0.00032598124 (7.968 sec)
INFO:tensorflow:global_step/sec: 12.5652
INFO:tensorflow:loss = 1.3361937, step = 20460 (7.956 sec)
INFO:tensorflow:lr = 0.0003256538 (7.955 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.18867
INFO:tensorflow:loss = 1.3309901, step = 20560 (12.216 sec)
INFO:tensorflow:lr = 0.00032532666 (12.223 sec)
INFO:tensorflow:global_step/sec: 12.6948
INFO:tensorflow:loss = 1.3054596, step = 20660 (7.873 sec)
INFO:tensorflow:lr = 0.00032499988 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.5301
INFO:tensorflow:loss = 1.2792804, step = 20760 (7.981 sec)
INFO:tensorflow:lr = 0.0003246734 (7.985 sec)
INFO:tensorflow:global_step/sec: 12.663
INFO:tensorflow:loss = 1.3026924, step = 20860 (7.897 sec)
INFO:tensorflow:lr = 0.00032434726 (7.895 sec)
INFO:tensorflow:global_step/sec: 12.7051
INFO:tensorflow:loss = 1.358359, step = 20960 (7.876 sec)
INFO:tensorflow:lr = 0.00032402144 (7.874 sec)
INFO:tensorflow:global_step/sec: 12.7735
INFO:tensorflow:loss = 1.3020449, step = 21060 (7.823 sec)
INFO:tensorflow:lr = 0.00032369597 (7.823 sec)
INFO:tensorflow:global_step/sec: 13.1674
INFO:tensorflow:loss = 1.3003097, step = 21160 (7.595 sec)
INFO:tensorflow:lr = 0.00032337077 (7.595 sec)
INFO:tensorflow:global_step/sec: 12.728
INFO:tensorflow:loss = 1.2887429, step = 21260 (7.863 sec)
INFO:tensorflow:lr = 0.00032304594 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.7174
INFO:tensorflow:loss = 1.3302351, step = 21360 (7.857 sec)
INFO:tensorflow:lr = 0.00032272143 (7.860 sec)
INFO:tensorflow:global_step/sec: 12.7107
INFO:tensorflow:loss = 1.3106179, step = 21460 (7.867 sec)
INFO:tensorflow:lr = 0.00032239725 (7.864 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.24138
INFO:tensorflow:loss = 1.2837156, step = 21560 (12.140 sec)
INFO:tensorflow:lr = 0.0003220734 (12.140 sec)
INFO:tensorflow:global_step/sec: 13.0831
INFO:tensorflow:loss = 1.3228761, step = 21660 (7.645 sec)
INFO:tensorflow:lr = 0.00032174986 (7.644 sec)
INFO:tensorflow:global_step/sec: 12.8614
INFO:tensorflow:loss = 1.312087, step = 21760 (7.773 sec)
INFO:tensorflow:lr = 0.00032142666 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.7587
INFO:tensorflow:loss = 1.3115535, step = 21860 (7.835 sec)
INFO:tensorflow:lr = 0.00032110378 (7.834 sec)
INFO:tensorflow:global_step/sec: 12.4595
INFO:tensorflow:loss = 1.2954695, step = 21960 (8.028 sec)
INFO:tensorflow:lr = 0.00032078126 (8.029 sec)
INFO:tensorflow:global_step/sec: 12.8623
INFO:tensorflow:loss = 1.2953963, step = 22060 (7.773 sec)
INFO:tensorflow:lr = 0.000320459 (7.774 sec)
INFO:tensorflow:global_step/sec: 12.7449
INFO:tensorflow:loss = 1.3128618, step = 22160 (7.843 sec)
INFO:tensorflow:lr = 0.00032013707 (7.847 sec)
INFO:tensorflow:global_step/sec: 12.622
INFO:tensorflow:loss = 1.3067259, step = 22260 (7.925 sec)
INFO:tensorflow:lr = 0.00031981547 (7.921 sec)
INFO:tensorflow:global_step/sec: 12.5084
INFO:tensorflow:loss = 1.3142986, step = 22360 (7.993 sec)
INFO:tensorflow:lr = 0.00031949423 (7.993 sec)
INFO:tensorflow:global_step/sec: 12.831
INFO:tensorflow:loss = 1.3074307, step = 22460 (7.793 sec)
INFO:tensorflow:lr = 0.00031917327 (7.792 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.2055
INFO:tensorflow:loss = 1.2916995, step = 22560 (12.186 sec)
INFO:tensorflow:lr = 0.0003188527 (12.189 sec)
INFO:tensorflow:global_step/sec: 12.8634
INFO:tensorflow:loss = 1.2907739, step = 22660 (7.779 sec)
INFO:tensorflow:lr = 0.00031853237 (7.777 sec)
INFO:tensorflow:global_step/sec: 12.8974
INFO:tensorflow:loss = 1.2921684, step = 22760 (7.755 sec)
INFO:tensorflow:lr = 0.0003182124 (7.759 sec)
INFO:tensorflow:global_step/sec: 12.7443
INFO:tensorflow:loss = 1.2923056, step = 22860 (7.844 sec)
INFO:tensorflow:lr = 0.00031789276 (7.840 sec)
INFO:tensorflow:global_step/sec: 12.6921
INFO:tensorflow:loss = 1.2913071, step = 22960 (7.875 sec)
INFO:tensorflow:lr = 0.00031757343 (7.875 sec)
INFO:tensorflow:global_step/sec: 12.6585
INFO:tensorflow:loss = 1.2972944, step = 23060 (7.902 sec)
INFO:tensorflow:lr = 0.0003172544 (7.903 sec)
INFO:tensorflow:global_step/sec: 12.8911
INFO:tensorflow:loss = 1.28654, step = 23160 (7.756 sec)
INFO:tensorflow:lr = 0.0003169357 (7.755 sec)
INFO:tensorflow:global_step/sec: 12.6817
INFO:tensorflow:loss = 1.3095835, step = 23260 (7.889 sec)
INFO:tensorflow:lr = 0.00031661734 (7.889 sec)
INFO:tensorflow:global_step/sec: 12.514
INFO:tensorflow:loss = 1.3168696, step = 23360 (7.989 sec)
INFO:tensorflow:lr = 0.0003162993 (7.989 sec)
INFO:tensorflow:global_step/sec: 12.7126
INFO:tensorflow:loss = 1.2910316, step = 23460 (7.864 sec)
INFO:tensorflow:lr = 0.00031598154 (7.863 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.40655
INFO:tensorflow:loss = 1.2932509, step = 23560 (11.898 sec)
INFO:tensorflow:lr = 0.00031566413 (11.898 sec)
INFO:tensorflow:global_step/sec: 12.81
INFO:tensorflow:loss = 1.2906524, step = 23660 (7.808 sec)
INFO:tensorflow:lr = 0.00031534708 (7.809 sec)
INFO:tensorflow:global_step/sec: 13.0227
INFO:tensorflow:loss = 1.2930197, step = 23760 (7.675 sec)
INFO:tensorflow:lr = 0.00031503028 (7.678 sec)
INFO:tensorflow:global_step/sec: 12.5006
INFO:tensorflow:loss = 1.3265917, step = 23860 (7.999 sec)
INFO:tensorflow:lr = 0.0003147138 (7.995 sec)
INFO:tensorflow:global_step/sec: 12.707
INFO:tensorflow:loss = 1.2934457, step = 23960 (7.870 sec)
INFO:tensorflow:lr = 0.00031439765 (7.870 sec)
INFO:tensorflow:global_step/sec: 12.3301
INFO:tensorflow:loss = 1.303881, step = 24060 (8.110 sec)
INFO:tensorflow:lr = 0.00031408184 (8.110 sec)
INFO:tensorflow:global_step/sec: 12.8441
INFO:tensorflow:loss = 1.3020984, step = 24160 (7.789 sec)
INFO:tensorflow:lr = 0.00031376636 (7.790 sec)
INFO:tensorflow:global_step/sec: 12.703
INFO:tensorflow:loss = 1.3089213, step = 24260 (7.869 sec)
INFO:tensorflow:lr = 0.00031345116 (7.871 sec)
INFO:tensorflow:global_step/sec: 12.6076
INFO:tensorflow:loss = 1.2835792, step = 24360 (7.932 sec)
INFO:tensorflow:lr = 0.0003131363 (7.930 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.6078
INFO:tensorflow:loss = 1.30011, step = 24460 (11.617 sec)
INFO:tensorflow:lr = 0.00031282174 (11.619 sec)
INFO:tensorflow:global_step/sec: 12.8552
INFO:tensorflow:loss = 1.2753806, step = 24560 (7.786 sec)
INFO:tensorflow:lr = 0.0003125075 (7.783 sec)
INFO:tensorflow:global_step/sec: 12.3914
INFO:tensorflow:loss = 1.2902484, step = 24660 (8.063 sec)
INFO:tensorflow:lr = 0.0003121936 (8.063 sec)
INFO:tensorflow:global_step/sec: 12.7618
INFO:tensorflow:loss = 1.3047645, step = 24760 (7.836 sec)
INFO:tensorflow:lr = 0.00031188 (7.837 sec)
INFO:tensorflow:global_step/sec: 12.8865
INFO:tensorflow:loss = 1.2961329, step = 24860 (7.767 sec)
INFO:tensorflow:lr = 0.0003115667 (7.767 sec)
INFO:tensorflow:global_step/sec: 12.9008
INFO:tensorflow:loss = 1.2774085, step = 24960 (7.748 sec)
INFO:tensorflow:lr = 0.0003112537 (7.749 sec)
INFO:tensorflow:global_step/sec: 12.804
INFO:tensorflow:loss = 1.2744296, step = 25060 (7.810 sec)
INFO:tensorflow:lr = 0.00031094105 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.8994
INFO:tensorflow:loss = 1.299216, step = 25160 (7.749 sec)
INFO:tensorflow:lr = 0.0003106287 (7.748 sec)
INFO:tensorflow:global_step/sec: 12.8665
INFO:tensorflow:loss = 1.2802867, step = 25260 (7.776 sec)
INFO:tensorflow:lr = 0.00031031665 (7.777 sec)
INFO:tensorflow:global_step/sec: 12.7289
INFO:tensorflow:loss = 1.3002493, step = 25360 (7.857 sec)
INFO:tensorflow:lr = 0.00031000495 (7.855 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.09796
INFO:tensorflow:loss = 1.315465, step = 25460 (10.988 sec)
INFO:tensorflow:lr = 0.00030969354 (10.988 sec)
INFO:tensorflow:global_step/sec: 12.8865
INFO:tensorflow:loss = 1.2996613, step = 25560 (7.759 sec)
INFO:tensorflow:lr = 0.00030938242 (7.765 sec)
INFO:tensorflow:global_step/sec: 12.6831
INFO:tensorflow:loss = 1.2841055, step = 25660 (7.888 sec)
INFO:tensorflow:lr = 0.00030907162 (7.883 sec)
INFO:tensorflow:global_step/sec: 12.9739
INFO:tensorflow:loss = 1.2864649, step = 25760 (7.708 sec)
INFO:tensorflow:lr = 0.00030876117 (7.708 sec)
INFO:tensorflow:global_step/sec: 12.7143
INFO:tensorflow:loss = 1.2838994, step = 25860 (7.866 sec)
INFO:tensorflow:lr = 0.000308451 (7.866 sec)
INFO:tensorflow:global_step/sec: 13.0226
INFO:tensorflow:loss = 1.2848786, step = 25960 (7.674 sec)
INFO:tensorflow:lr = 0.00030814117 (7.675 sec)
INFO:tensorflow:global_step/sec: 12.8533
INFO:tensorflow:loss = 1.2814575, step = 26060 (7.780 sec)
INFO:tensorflow:lr = 0.00030783165 (7.781 sec)
INFO:tensorflow:global_step/sec: 12.5923
INFO:tensorflow:loss = 1.2896702, step = 26160 (7.941 sec)
INFO:tensorflow:lr = 0.0003075224 (7.943 sec)
INFO:tensorflow:global_step/sec: 12.7118
INFO:tensorflow:loss = 1.2936949, step = 26260 (7.871 sec)
INFO:tensorflow:lr = 0.00030721352 (7.868 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1137
INFO:tensorflow:loss = 1.2839646, step = 26360 (8.256 sec)
INFO:tensorflow:lr = 0.0003069049 (8.256 sec)
INFO:tensorflow:global_step/sec: 10.4227
INFO:tensorflow:loss = 1.2964764, step = 26460 (9.589 sec)
INFO:tensorflow:lr = 0.0003065966 (9.591 sec)
INFO:tensorflow:global_step/sec: 12.6315
INFO:tensorflow:loss = 1.2799263, step = 26560 (7.921 sec)
INFO:tensorflow:lr = 0.0003062886 (7.919 sec)
INFO:tensorflow:global_step/sec: 12.813
INFO:tensorflow:loss = 1.2930382, step = 26660 (7.801 sec)
INFO:tensorflow:lr = 0.00030598094 (7.802 sec)
INFO:tensorflow:global_step/sec: 12.7167
INFO:tensorflow:loss = 1.2775872, step = 26760 (7.863 sec)
INFO:tensorflow:lr = 0.00030567357 (7.862 sec)
INFO:tensorflow:global_step/sec: 12.5424
INFO:tensorflow:loss = 1.2938052, step = 26860 (7.977 sec)
INFO:tensorflow:lr = 0.0003053665 (7.977 sec)
INFO:tensorflow:global_step/sec: 12.9605
INFO:tensorflow:loss = 1.3075194, step = 26960 (7.711 sec)
INFO:tensorflow:lr = 0.00030505977 (7.711 sec)
INFO:tensorflow:global_step/sec: 12.644
INFO:tensorflow:loss = 1.2937547, step = 27060 (7.909 sec)
INFO:tensorflow:lr = 0.0003047533 (7.910 sec)
INFO:tensorflow:global_step/sec: 13.0595
INFO:tensorflow:loss = 1.3111972, step = 27160 (7.662 sec)
INFO:tensorflow:lr = 0.0003044472 (7.662 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.5976
INFO:tensorflow:loss = 1.2813871, step = 27260 (7.937 sec)
INFO:tensorflow:lr = 0.00030414137 (7.938 sec)
INFO:tensorflow:global_step/sec: 10.8542
INFO:tensorflow:loss = 1.2741573, step = 27360 (9.215 sec)
INFO:tensorflow:lr = 0.00030383587 (9.214 sec)
INFO:tensorflow:global_step/sec: 12.9826
INFO:tensorflow:loss = 1.2877482, step = 27460 (7.701 sec)
INFO:tensorflow:lr = 0.00030353063 (7.701 sec)
INFO:tensorflow:global_step/sec: 12.439
INFO:tensorflow:loss = 1.2878534, step = 27560 (8.035 sec)
INFO:tensorflow:lr = 0.0003032257 (8.036 sec)
INFO:tensorflow:global_step/sec: 12.477
INFO:tensorflow:loss = 1.2885484, step = 27660 (8.014 sec)
INFO:tensorflow:lr = 0.0003029211 (8.013 sec)
INFO:tensorflow:global_step/sec: 12.4905
INFO:tensorflow:loss = 1.2796994, step = 27760 (8.006 sec)
INFO:tensorflow:lr = 0.00030261683 (8.006 sec)
INFO:tensorflow:global_step/sec: 12.8284
INFO:tensorflow:loss = 1.2923272, step = 27860 (7.799 sec)
INFO:tensorflow:lr = 0.00030231284 (7.802 sec)
INFO:tensorflow:global_step/sec: 12.7148
INFO:tensorflow:loss = 1.2721691, step = 27960 (7.866 sec)
INFO:tensorflow:lr = 0.00030200917 (7.864 sec)
INFO:tensorflow:global_step/sec: 12.9207
INFO:tensorflow:loss = 1.2802061, step = 28060 (7.739 sec)
INFO:tensorflow:lr = 0.0003017058 (7.738 sec)
INFO:tensorflow:global_step/sec: 12.7999
INFO:tensorflow:loss = 1.2846918, step = 28160 (7.813 sec)
INFO:tensorflow:lr = 0.00030140273 (7.813 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.8178
INFO:tensorflow:loss = 1.2896069, step = 28260 (8.461 sec)
INFO:tensorflow:lr = 0.00030109996 (8.461 sec)
INFO:tensorflow:global_step/sec: 11.4031
INFO:tensorflow:loss = 1.2732713, step = 28360 (8.769 sec)
INFO:tensorflow:lr = 0.0003007975 (8.769 sec)
INFO:tensorflow:global_step/sec: 12.6608
INFO:tensorflow:loss = 1.2760838, step = 28460 (7.895 sec)
INFO:tensorflow:lr = 0.00030049533 (7.900 sec)
INFO:tensorflow:global_step/sec: 12.5912
INFO:tensorflow:loss = 1.285021, step = 28560 (7.944 sec)
INFO:tensorflow:lr = 0.00030019347 (7.939 sec)
INFO:tensorflow:global_step/sec: 12.6283
INFO:tensorflow:loss = 1.2831901, step = 28660 (7.922 sec)
INFO:tensorflow:lr = 0.00029989192 (7.922 sec)
INFO:tensorflow:global_step/sec: 12.5436
INFO:tensorflow:loss = 1.2798933, step = 28760 (7.970 sec)
INFO:tensorflow:lr = 0.00029959064 (7.970 sec)
INFO:tensorflow:global_step/sec: 12.9383
INFO:tensorflow:loss = 1.2905648, step = 28860 (7.730 sec)
INFO:tensorflow:lr = 0.0002992897 (7.730 sec)
INFO:tensorflow:global_step/sec: 13.0137
INFO:tensorflow:loss = 1.2883794, step = 28960 (7.680 sec)
INFO:tensorflow:lr = 0.00029898906 (7.680 sec)
INFO:tensorflow:global_step/sec: 12.8798
INFO:tensorflow:loss = 1.2834778, step = 29060 (7.764 sec)
INFO:tensorflow:lr = 0.00029868874 (7.764 sec)
INFO:tensorflow:global_step/sec: 12.8779
INFO:tensorflow:loss = 1.2783784, step = 29160 (7.765 sec)
INFO:tensorflow:lr = 0.00029838865 (7.770 sec)
INFO:tensorflow:global_step/sec: 12.7015
INFO:tensorflow:loss = 1.2855964, step = 29260 (7.874 sec)
INFO:tensorflow:lr = 0.00029808894 (7.877 sec)
INFO:tensorflow:Saving checkpoints for 29340 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.302846.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-29340
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-29340
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.702
INFO:tensorflow:Best (Exact Match) Accuracy: 0.702
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-29340
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 29340 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2902774, step = 29340
INFO:tensorflow:lr = 0.0002978494
INFO:tensorflow:global_step/sec: 10.3227
INFO:tensorflow:loss = 1.3032885, step = 29440 (9.692 sec)
INFO:tensorflow:lr = 0.0002975502 (9.693 sec)
INFO:tensorflow:global_step/sec: 12.3379
INFO:tensorflow:loss = 1.2751992, step = 29540 (8.106 sec)
INFO:tensorflow:lr = 0.0002972513 (8.105 sec)
INFO:tensorflow:global_step/sec: 12.8195
INFO:tensorflow:loss = 1.2912132, step = 29640 (7.800 sec)
INFO:tensorflow:lr = 0.0002969527 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.4392
INFO:tensorflow:loss = 1.2732567, step = 29740 (8.036 sec)
INFO:tensorflow:lr = 0.00029665435 (8.039 sec)
INFO:tensorflow:global_step/sec: 12.5569
INFO:tensorflow:loss = 1.2769362, step = 29840 (7.967 sec)
INFO:tensorflow:lr = 0.0002963564 (7.966 sec)
INFO:tensorflow:global_step/sec: 12.9836
INFO:tensorflow:loss = 1.2793033, step = 29940 (7.699 sec)
INFO:tensorflow:lr = 0.00029605866 (7.702 sec)
INFO:tensorflow:global_step/sec: 12.5844
INFO:tensorflow:loss = 1.2748176, step = 30040 (7.947 sec)
INFO:tensorflow:lr = 0.0002957613 (7.943 sec)
INFO:tensorflow:global_step/sec: 12.7121
INFO:tensorflow:loss = 1.2927984, step = 30140 (7.866 sec)
INFO:tensorflow:lr = 0.00029546418 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.6349
INFO:tensorflow:loss = 1.2704685, step = 30240 (7.915 sec)
INFO:tensorflow:lr = 0.00029516738 (7.913 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.30175
INFO:tensorflow:loss = 1.2709056, step = 30340 (12.049 sec)
INFO:tensorflow:lr = 0.0002948709 (12.050 sec)
INFO:tensorflow:global_step/sec: 12.6638
INFO:tensorflow:loss = 1.2773079, step = 30440 (7.898 sec)
INFO:tensorflow:lr = 0.00029457468 (7.897 sec)
INFO:tensorflow:global_step/sec: 12.4741
INFO:tensorflow:loss = 1.2764441, step = 30540 (8.016 sec)
INFO:tensorflow:lr = 0.00029427878 (8.018 sec)
INFO:tensorflow:global_step/sec: 12.7293
INFO:tensorflow:loss = 1.2704064, step = 30640 (7.857 sec)
INFO:tensorflow:lr = 0.00029398318 (7.855 sec)
INFO:tensorflow:global_step/sec: 12.6686
INFO:tensorflow:loss = 1.2712173, step = 30740 (7.893 sec)
INFO:tensorflow:lr = 0.00029368783 (7.893 sec)
INFO:tensorflow:global_step/sec: 12.7314
INFO:tensorflow:loss = 1.3050127, step = 30840 (7.851 sec)
INFO:tensorflow:lr = 0.0002933928 (7.852 sec)
INFO:tensorflow:global_step/sec: 12.8834
INFO:tensorflow:loss = 1.2784503, step = 30940 (7.765 sec)
INFO:tensorflow:lr = 0.0002930981 (7.764 sec)
INFO:tensorflow:global_step/sec: 12.88
INFO:tensorflow:loss = 1.2874303, step = 31040 (7.764 sec)
INFO:tensorflow:lr = 0.00029280368 (7.764 sec)
INFO:tensorflow:global_step/sec: 12.9115
INFO:tensorflow:loss = 1.2781101, step = 31140 (7.746 sec)
INFO:tensorflow:lr = 0.00029250956 (7.746 sec)
INFO:tensorflow:global_step/sec: 12.6392
INFO:tensorflow:loss = 1.2868813, step = 31240 (7.911 sec)
INFO:tensorflow:lr = 0.00029221573 (7.913 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.37144
INFO:tensorflow:loss = 1.2743399, step = 31340 (11.942 sec)
INFO:tensorflow:lr = 0.0002919222 (11.940 sec)
INFO:tensorflow:global_step/sec: 12.4837
INFO:tensorflow:loss = 1.2944267, step = 31440 (8.010 sec)
INFO:tensorflow:lr = 0.00029162894 (8.011 sec)
INFO:tensorflow:global_step/sec: 12.7212
INFO:tensorflow:loss = 1.3041955, step = 31540 (7.865 sec)
INFO:tensorflow:lr = 0.000291336 (7.864 sec)
INFO:tensorflow:global_step/sec: 12.9116
INFO:tensorflow:loss = 1.27363, step = 31640 (7.741 sec)
INFO:tensorflow:lr = 0.00029104334 (7.742 sec)
INFO:tensorflow:global_step/sec: 12.7415
INFO:tensorflow:loss = 1.3110542, step = 31740 (7.853 sec)
INFO:tensorflow:lr = 0.00029075096 (7.852 sec)
INFO:tensorflow:global_step/sec: 13.0078
INFO:tensorflow:loss = 1.2707573, step = 31840 (7.687 sec)
INFO:tensorflow:lr = 0.00029045888 (7.686 sec)
INFO:tensorflow:global_step/sec: 12.6669
INFO:tensorflow:loss = 1.2790074, step = 31940 (7.896 sec)
INFO:tensorflow:lr = 0.00029016714 (7.896 sec)
INFO:tensorflow:global_step/sec: 12.9049
INFO:tensorflow:loss = 1.27734, step = 32040 (7.748 sec)
INFO:tensorflow:lr = 0.00028987564 (7.748 sec)
INFO:tensorflow:global_step/sec: 12.6828
INFO:tensorflow:loss = 1.2766548, step = 32140 (7.881 sec)
INFO:tensorflow:lr = 0.00028958445 (7.881 sec)
INFO:tensorflow:global_step/sec: 12.6536
INFO:tensorflow:loss = 1.270347, step = 32240 (7.904 sec)
INFO:tensorflow:lr = 0.00028929356 (7.903 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.34015
INFO:tensorflow:loss = 1.2706623, step = 32340 (11.991 sec)
INFO:tensorflow:lr = 0.00028900296 (11.991 sec)
INFO:tensorflow:global_step/sec: 12.6105
INFO:tensorflow:loss = 1.2677004, step = 32440 (7.932 sec)
INFO:tensorflow:lr = 0.00028871265 (7.933 sec)
INFO:tensorflow:global_step/sec: 12.8231
INFO:tensorflow:loss = 1.2773702, step = 32540 (7.795 sec)
INFO:tensorflow:lr = 0.00028842266 (7.799 sec)
INFO:tensorflow:global_step/sec: 12.9974
INFO:tensorflow:loss = 1.2898613, step = 32640 (7.699 sec)
INFO:tensorflow:lr = 0.0002881329 (7.695 sec)
INFO:tensorflow:global_step/sec: 12.4858
INFO:tensorflow:loss = 1.2836853, step = 32740 (8.008 sec)
INFO:tensorflow:lr = 0.00028784346 (8.008 sec)
INFO:tensorflow:global_step/sec: 12.8931
INFO:tensorflow:loss = 1.2881771, step = 32840 (7.751 sec)
INFO:tensorflow:lr = 0.00028755434 (7.752 sec)
INFO:tensorflow:global_step/sec: 12.819
INFO:tensorflow:loss = 1.276155, step = 32940 (7.801 sec)
INFO:tensorflow:lr = 0.00028726546 (7.801 sec)
INFO:tensorflow:global_step/sec: 12.7679
INFO:tensorflow:loss = 1.3059559, step = 33040 (7.833 sec)
INFO:tensorflow:lr = 0.0002869769 (7.832 sec)
INFO:tensorflow:global_step/sec: 12.8372
INFO:tensorflow:loss = 1.2818044, step = 33140 (7.793 sec)
INFO:tensorflow:lr = 0.00028668862 (7.793 sec)
INFO:tensorflow:global_step/sec: 12.6358
INFO:tensorflow:loss = 1.272675, step = 33240 (7.911 sec)
INFO:tensorflow:lr = 0.00028640064 (7.910 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.43506
INFO:tensorflow:loss = 1.2660692, step = 33340 (11.861 sec)
INFO:tensorflow:lr = 0.00028611295 (11.861 sec)
INFO:tensorflow:global_step/sec: 12.8342
INFO:tensorflow:loss = 1.274544, step = 33440 (7.791 sec)
INFO:tensorflow:lr = 0.00028582552 (7.792 sec)
INFO:tensorflow:global_step/sec: 12.4115
INFO:tensorflow:loss = 1.282098, step = 33540 (8.055 sec)
INFO:tensorflow:lr = 0.0002855384 (8.054 sec)
INFO:tensorflow:global_step/sec: 12.8384
INFO:tensorflow:loss = 1.271896, step = 33640 (7.786 sec)
INFO:tensorflow:lr = 0.0002852516 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.7124
INFO:tensorflow:loss = 1.2689635, step = 33740 (7.871 sec)
INFO:tensorflow:lr = 0.00028496503 (7.870 sec)
INFO:tensorflow:global_step/sec: 12.8111
INFO:tensorflow:loss = 1.2838228, step = 33840 (7.802 sec)
INFO:tensorflow:lr = 0.0002846788 (7.802 sec)
INFO:tensorflow:global_step/sec: 12.779
INFO:tensorflow:loss = 1.2770149, step = 33940 (7.825 sec)
INFO:tensorflow:lr = 0.0002843928 (7.825 sec)
INFO:tensorflow:global_step/sec: 13.0904
INFO:tensorflow:loss = 1.266436, step = 34040 (7.638 sec)
INFO:tensorflow:lr = 0.00028410714 (7.642 sec)
INFO:tensorflow:global_step/sec: 12.6556
INFO:tensorflow:loss = 1.2893338, step = 34140 (7.908 sec)
INFO:tensorflow:lr = 0.00028382175 (7.904 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.60813
INFO:tensorflow:loss = 1.2691966, step = 34240 (11.615 sec)
INFO:tensorflow:lr = 0.00028353665 (11.616 sec)
INFO:tensorflow:global_step/sec: 12.5767
INFO:tensorflow:loss = 1.2772423, step = 34340 (7.948 sec)
INFO:tensorflow:lr = 0.0002832518 (7.948 sec)
INFO:tensorflow:global_step/sec: 12.8296
INFO:tensorflow:loss = 1.2685624, step = 34440 (7.794 sec)
INFO:tensorflow:lr = 0.00028296726 (7.794 sec)
INFO:tensorflow:global_step/sec: 12.8444
INFO:tensorflow:loss = 1.2796193, step = 34540 (7.787 sec)
INFO:tensorflow:lr = 0.000282683 (7.786 sec)
INFO:tensorflow:global_step/sec: 12.9569
INFO:tensorflow:loss = 1.2901658, step = 34640 (7.717 sec)
INFO:tensorflow:lr = 0.00028239907 (7.717 sec)
INFO:tensorflow:global_step/sec: 12.907
INFO:tensorflow:loss = 1.2649109, step = 34740 (7.754 sec)
INFO:tensorflow:lr = 0.0002821154 (7.754 sec)
INFO:tensorflow:global_step/sec: 12.8117
INFO:tensorflow:loss = 1.2722139, step = 34840 (7.804 sec)
INFO:tensorflow:lr = 0.000281832 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.7117
INFO:tensorflow:loss = 1.2853813, step = 34940 (7.862 sec)
INFO:tensorflow:lr = 0.00028154888 (7.861 sec)
INFO:tensorflow:global_step/sec: 12.6376
INFO:tensorflow:loss = 1.2778032, step = 35040 (7.915 sec)
INFO:tensorflow:lr = 0.00028126605 (7.916 sec)
INFO:tensorflow:global_step/sec: 12.8204
INFO:tensorflow:loss = 1.2695035, step = 35140 (7.799 sec)
INFO:tensorflow:lr = 0.00028098354 (7.798 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.08278
INFO:tensorflow:loss = 1.2864376, step = 35240 (11.012 sec)
INFO:tensorflow:lr = 0.00028070126 (11.012 sec)
INFO:tensorflow:global_step/sec: 12.3979
INFO:tensorflow:loss = 1.2693602, step = 35340 (8.065 sec)
INFO:tensorflow:lr = 0.0002804193 (8.068 sec)
INFO:tensorflow:global_step/sec: 12.9214
INFO:tensorflow:loss = 1.2835609, step = 35440 (7.743 sec)
INFO:tensorflow:lr = 0.0002801376 (7.740 sec)
INFO:tensorflow:global_step/sec: 12.5272
INFO:tensorflow:loss = 1.2637956, step = 35540 (7.981 sec)
INFO:tensorflow:lr = 0.0002798562 (7.981 sec)
INFO:tensorflow:global_step/sec: 12.9356
INFO:tensorflow:loss = 1.2700819, step = 35640 (7.730 sec)
INFO:tensorflow:lr = 0.00027957506 (7.730 sec)
INFO:tensorflow:global_step/sec: 12.8449
INFO:tensorflow:loss = 1.2769866, step = 35740 (7.781 sec)
INFO:tensorflow:lr = 0.0002792942 (7.781 sec)
INFO:tensorflow:global_step/sec: 12.7669
INFO:tensorflow:loss = 1.2878106, step = 35840 (7.833 sec)
INFO:tensorflow:lr = 0.00027901368 (7.834 sec)
INFO:tensorflow:global_step/sec: 12.4951
INFO:tensorflow:loss = 1.2848315, step = 35940 (8.003 sec)
INFO:tensorflow:lr = 0.00027873338 (8.003 sec)
INFO:tensorflow:global_step/sec: 12.9564
INFO:tensorflow:loss = 1.2649281, step = 36040 (7.718 sec)
INFO:tensorflow:lr = 0.0002784534 (7.717 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.5196
INFO:tensorflow:loss = 1.2739855, step = 36140 (7.988 sec)
INFO:tensorflow:lr = 0.00027817365 (7.988 sec)
INFO:tensorflow:global_step/sec: 10.5621
INFO:tensorflow:loss = 1.2650766, step = 36240 (9.467 sec)
INFO:tensorflow:lr = 0.00027789423 (9.468 sec)
INFO:tensorflow:global_step/sec: 12.7425
INFO:tensorflow:loss = 1.2738944, step = 36340 (7.853 sec)
INFO:tensorflow:lr = 0.0002776151 (7.853 sec)
INFO:tensorflow:global_step/sec: 12.5367
INFO:tensorflow:loss = 1.2752404, step = 36440 (7.974 sec)
INFO:tensorflow:lr = 0.00027733622 (7.974 sec)
INFO:tensorflow:global_step/sec: 12.9389
INFO:tensorflow:loss = 1.2804888, step = 36540 (7.727 sec)
INFO:tensorflow:lr = 0.00027705764 (7.726 sec)
INFO:tensorflow:global_step/sec: 12.625
INFO:tensorflow:loss = 1.2651876, step = 36640 (7.925 sec)
INFO:tensorflow:lr = 0.00027677935 (7.925 sec)
INFO:tensorflow:global_step/sec: 12.6561
INFO:tensorflow:loss = 1.2666996, step = 36740 (7.898 sec)
INFO:tensorflow:lr = 0.0002765013 (7.899 sec)
INFO:tensorflow:global_step/sec: 12.6222
INFO:tensorflow:loss = 1.288384, step = 36840 (7.924 sec)
INFO:tensorflow:lr = 0.00027622355 (7.923 sec)
INFO:tensorflow:global_step/sec: 12.5421
INFO:tensorflow:loss = 1.2701671, step = 36940 (7.970 sec)
INFO:tensorflow:lr = 0.00027594608 (7.972 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 13.1055
INFO:tensorflow:loss = 1.2687078, step = 37040 (7.635 sec)
INFO:tensorflow:lr = 0.00027566886 (7.636 sec)
INFO:tensorflow:global_step/sec: 11.2215
INFO:tensorflow:loss = 1.2759229, step = 37140 (8.909 sec)
INFO:tensorflow:lr = 0.00027539194 (8.908 sec)
INFO:tensorflow:global_step/sec: 12.6468
INFO:tensorflow:loss = 1.2758858, step = 37240 (7.912 sec)
INFO:tensorflow:lr = 0.0002751153 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.3398
INFO:tensorflow:loss = 1.266111, step = 37340 (8.101 sec)
INFO:tensorflow:lr = 0.00027483894 (8.101 sec)
INFO:tensorflow:global_step/sec: 12.7537
INFO:tensorflow:loss = 1.2725819, step = 37440 (7.836 sec)
INFO:tensorflow:lr = 0.00027456286 (7.837 sec)
INFO:tensorflow:global_step/sec: 13.1806
INFO:tensorflow:loss = 1.2705429, step = 37540 (7.588 sec)
INFO:tensorflow:lr = 0.00027428704 (7.586 sec)
INFO:tensorflow:global_step/sec: 13.1254
INFO:tensorflow:loss = 1.2754468, step = 37640 (7.624 sec)
INFO:tensorflow:lr = 0.00027401155 (7.625 sec)
INFO:tensorflow:global_step/sec: 12.63
INFO:tensorflow:loss = 1.2782013, step = 37740 (7.914 sec)
INFO:tensorflow:lr = 0.00027373628 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.6325
INFO:tensorflow:loss = 1.2712921, step = 37840 (7.913 sec)
INFO:tensorflow:lr = 0.0002734613 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.4989
INFO:tensorflow:loss = 1.2724797, step = 37940 (8.004 sec)
INFO:tensorflow:lr = 0.00027318663 (8.004 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.9641
INFO:tensorflow:loss = 1.2724507, step = 38040 (8.361 sec)
INFO:tensorflow:lr = 0.0002729122 (8.360 sec)
INFO:tensorflow:global_step/sec: 11.1246
INFO:tensorflow:loss = 1.2643977, step = 38140 (8.983 sec)
INFO:tensorflow:lr = 0.00027263802 (8.984 sec)
INFO:tensorflow:global_step/sec: 12.8127
INFO:tensorflow:loss = 1.2717063, step = 38240 (7.812 sec)
INFO:tensorflow:lr = 0.00027236415 (7.811 sec)
INFO:tensorflow:global_step/sec: 12.9284
INFO:tensorflow:loss = 1.2676791, step = 38340 (7.731 sec)
INFO:tensorflow:lr = 0.00027209055 (7.738 sec)
INFO:tensorflow:global_step/sec: 13.087
INFO:tensorflow:loss = 1.2721039, step = 38440 (7.644 sec)
INFO:tensorflow:lr = 0.00027181726 (7.637 sec)
INFO:tensorflow:global_step/sec: 12.8318
INFO:tensorflow:loss = 1.2676044, step = 38540 (7.788 sec)
INFO:tensorflow:lr = 0.00027154418 (7.789 sec)
INFO:tensorflow:global_step/sec: 12.6748
INFO:tensorflow:loss = 1.2753915, step = 38640 (7.894 sec)
INFO:tensorflow:lr = 0.00027127142 (7.894 sec)
INFO:tensorflow:global_step/sec: 12.6085
INFO:tensorflow:loss = 1.2720599, step = 38740 (7.926 sec)
INFO:tensorflow:lr = 0.00027099892 (7.926 sec)
INFO:tensorflow:global_step/sec: 12.7104
INFO:tensorflow:loss = 1.2626356, step = 38840 (7.874 sec)
INFO:tensorflow:lr = 0.0002707267 (7.874 sec)
INFO:tensorflow:global_step/sec: 12.696
INFO:tensorflow:loss = 1.2689277, step = 38940 (7.874 sec)
INFO:tensorflow:lr = 0.00027045477 (7.877 sec)
INFO:tensorflow:global_step/sec: 12.6811
INFO:tensorflow:loss = 1.2729962, step = 39040 (7.887 sec)
INFO:tensorflow:lr = 0.00027018308 (7.883 sec)
INFO:tensorflow:Saving checkpoints for 39120 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2890505.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-39120
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|_________________ | |
what times are playing the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-39120
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.707
INFO:tensorflow:Best (Exact Match) Accuracy: 0.707
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-39120
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 39120 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2692995, step = 39120
INFO:tensorflow:lr = 0.00026996594
INFO:tensorflow:global_step/sec: 10.423
INFO:tensorflow:loss = 1.2852095, step = 39220 (9.595 sec)
INFO:tensorflow:lr = 0.00026969475 (9.595 sec)
INFO:tensorflow:global_step/sec: 12.7697
INFO:tensorflow:loss = 1.2648714, step = 39320 (7.838 sec)
INFO:tensorflow:lr = 0.00026942385 (7.839 sec)
INFO:tensorflow:global_step/sec: 12.4705
INFO:tensorflow:loss = 1.263199, step = 39420 (8.017 sec)
INFO:tensorflow:lr = 0.00026915318 (8.017 sec)
INFO:tensorflow:global_step/sec: 12.5405
INFO:tensorflow:loss = 1.2712208, step = 39520 (7.972 sec)
INFO:tensorflow:lr = 0.0002688828 (7.972 sec)
INFO:tensorflow:global_step/sec: 12.4456
INFO:tensorflow:loss = 1.2705394, step = 39620 (8.033 sec)
INFO:tensorflow:lr = 0.0002686127 (8.033 sec)
INFO:tensorflow:global_step/sec: 12.6205
INFO:tensorflow:loss = 1.271941, step = 39720 (7.926 sec)
INFO:tensorflow:lr = 0.00026834288 (7.925 sec)
INFO:tensorflow:global_step/sec: 12.7271
INFO:tensorflow:loss = 1.2671064, step = 39820 (7.855 sec)
INFO:tensorflow:lr = 0.00026807332 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.8522
INFO:tensorflow:loss = 1.2827927, step = 39920 (7.787 sec)
INFO:tensorflow:lr = 0.00026780402 (7.787 sec)
INFO:tensorflow:global_step/sec: 13.0029
INFO:tensorflow:loss = 1.2829088, step = 40020 (7.690 sec)
INFO:tensorflow:lr = 0.000267535 (7.690 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.24724
INFO:tensorflow:loss = 1.2646608, step = 40120 (12.122 sec)
INFO:tensorflow:lr = 0.00026726627 (12.124 sec)
INFO:tensorflow:global_step/sec: 12.9064
INFO:tensorflow:loss = 1.2660892, step = 40220 (7.746 sec)
INFO:tensorflow:lr = 0.00026699778 (7.744 sec)
INFO:tensorflow:global_step/sec: 12.6494
INFO:tensorflow:loss = 1.2711158, step = 40320 (7.913 sec)
INFO:tensorflow:lr = 0.00026672956 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.736
INFO:tensorflow:loss = 1.2661604, step = 40420 (7.846 sec)
INFO:tensorflow:lr = 0.00026646163 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.5546
INFO:tensorflow:loss = 1.2606105, step = 40520 (7.964 sec)
INFO:tensorflow:lr = 0.00026619397 (7.961 sec)
INFO:tensorflow:global_step/sec: 12.7677
INFO:tensorflow:loss = 1.27701, step = 40620 (7.831 sec)
INFO:tensorflow:lr = 0.00026592656 (7.831 sec)
INFO:tensorflow:global_step/sec: 12.938
INFO:tensorflow:loss = 1.2607424, step = 40720 (7.732 sec)
INFO:tensorflow:lr = 0.00026565944 (7.733 sec)
INFO:tensorflow:global_step/sec: 12.8514
INFO:tensorflow:loss = 1.270835, step = 40820 (7.782 sec)
INFO:tensorflow:lr = 0.0002653926 (7.781 sec)
INFO:tensorflow:global_step/sec: 12.6607
INFO:tensorflow:loss = 1.2720766, step = 40920 (7.900 sec)
INFO:tensorflow:lr = 0.00026512597 (7.901 sec)
INFO:tensorflow:global_step/sec: 12.5758
INFO:tensorflow:loss = 1.2690309, step = 41020 (7.947 sec)
INFO:tensorflow:lr = 0.00026485967 (7.947 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.24971
INFO:tensorflow:loss = 1.2594911, step = 41120 (12.123 sec)
INFO:tensorflow:lr = 0.00026459363 (12.123 sec)
INFO:tensorflow:global_step/sec: 12.9862
INFO:tensorflow:loss = 1.283308, step = 41220 (7.699 sec)
INFO:tensorflow:lr = 0.00026432783 (7.700 sec)
INFO:tensorflow:global_step/sec: 12.7218
INFO:tensorflow:loss = 1.2755083, step = 41320 (7.865 sec)
INFO:tensorflow:lr = 0.00026406228 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.7322
INFO:tensorflow:loss = 1.268201, step = 41420 (7.849 sec)
INFO:tensorflow:lr = 0.000263797 (7.849 sec)
INFO:tensorflow:global_step/sec: 12.4848
INFO:tensorflow:loss = 1.2764183, step = 41520 (8.015 sec)
INFO:tensorflow:lr = 0.00026353204 (8.013 sec)
INFO:tensorflow:global_step/sec: 12.5466
INFO:tensorflow:loss = 1.2742758, step = 41620 (7.972 sec)
INFO:tensorflow:lr = 0.0002632673 (7.976 sec)
INFO:tensorflow:global_step/sec: 12.8135
INFO:tensorflow:loss = 1.2566481, step = 41720 (7.798 sec)
INFO:tensorflow:lr = 0.00026300285 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.9486
INFO:tensorflow:loss = 1.261928, step = 41820 (7.723 sec)
INFO:tensorflow:lr = 0.00026273864 (7.720 sec)
INFO:tensorflow:global_step/sec: 12.9607
INFO:tensorflow:loss = 1.2595774, step = 41920 (7.719 sec)
INFO:tensorflow:lr = 0.00026247473 (7.720 sec)
INFO:tensorflow:global_step/sec: 12.8198
INFO:tensorflow:loss = 1.2714173, step = 42020 (7.802 sec)
INFO:tensorflow:lr = 0.00026221108 (7.803 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.4317
INFO:tensorflow:loss = 1.2644148, step = 42120 (11.855 sec)
INFO:tensorflow:lr = 0.0002619477 (11.853 sec)
INFO:tensorflow:global_step/sec: 12.9926
INFO:tensorflow:loss = 1.2653011, step = 42220 (7.698 sec)
INFO:tensorflow:lr = 0.00026168456 (7.702 sec)
INFO:tensorflow:global_step/sec: 12.6806
INFO:tensorflow:loss = 1.2612933, step = 42320 (7.890 sec)
INFO:tensorflow:lr = 0.00026142164 (7.887 sec)
INFO:tensorflow:global_step/sec: 12.8074
INFO:tensorflow:loss = 1.2797445, step = 42420 (7.805 sec)
INFO:tensorflow:lr = 0.00026115906 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.6395
INFO:tensorflow:loss = 1.2628508, step = 42520 (7.913 sec)
INFO:tensorflow:lr = 0.00026089672 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.8445
INFO:tensorflow:loss = 1.2819867, step = 42620 (7.788 sec)
INFO:tensorflow:lr = 0.00026063464 (7.789 sec)
INFO:tensorflow:global_step/sec: 12.8711
INFO:tensorflow:loss = 1.2706252, step = 42720 (7.768 sec)
INFO:tensorflow:lr = 0.00026037282 (7.767 sec)
INFO:tensorflow:global_step/sec: 12.3352
INFO:tensorflow:loss = 1.2649037, step = 42820 (8.102 sec)
INFO:tensorflow:lr = 0.00026011126 (8.101 sec)
INFO:tensorflow:global_step/sec: 12.8112
INFO:tensorflow:loss = 1.2671568, step = 42920 (7.809 sec)
INFO:tensorflow:lr = 0.00025985 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.9254
INFO:tensorflow:loss = 1.2667608, step = 43020 (7.741 sec)
INFO:tensorflow:lr = 0.00025958894 (7.741 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.3895
INFO:tensorflow:loss = 1.2655711, step = 43120 (11.913 sec)
INFO:tensorflow:lr = 0.0002593282 (11.913 sec)
INFO:tensorflow:global_step/sec: 12.4018
INFO:tensorflow:loss = 1.267358, step = 43220 (8.066 sec)
INFO:tensorflow:lr = 0.0002590677 (8.069 sec)
INFO:tensorflow:global_step/sec: 12.6024
INFO:tensorflow:loss = 1.2700015, step = 43320 (7.935 sec)
INFO:tensorflow:lr = 0.00025880744 (7.934 sec)
INFO:tensorflow:global_step/sec: 12.5451
INFO:tensorflow:loss = 1.2641869, step = 43420 (7.972 sec)
INFO:tensorflow:lr = 0.00025854746 (7.972 sec)
INFO:tensorflow:global_step/sec: 12.8976
INFO:tensorflow:loss = 1.2638835, step = 43520 (7.755 sec)
INFO:tensorflow:lr = 0.00025828776 (7.753 sec)
INFO:tensorflow:global_step/sec: 12.6097
INFO:tensorflow:loss = 1.2561631, step = 43620 (7.927 sec)
INFO:tensorflow:lr = 0.0002580283 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.8381
INFO:tensorflow:loss = 1.2625241, step = 43720 (7.793 sec)
INFO:tensorflow:lr = 0.0002577691 (7.792 sec)
INFO:tensorflow:global_step/sec: 12.825
INFO:tensorflow:loss = 1.2693264, step = 43820 (7.798 sec)
INFO:tensorflow:lr = 0.00025751017 (7.798 sec)
INFO:tensorflow:global_step/sec: 12.7736
INFO:tensorflow:loss = 1.2730483, step = 43920 (7.822 sec)
INFO:tensorflow:lr = 0.0002572515 (7.823 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.75995
INFO:tensorflow:loss = 1.2587811, step = 44020 (11.418 sec)
INFO:tensorflow:lr = 0.00025699308 (11.420 sec)
INFO:tensorflow:global_step/sec: 12.8387
INFO:tensorflow:loss = 1.2614508, step = 44120 (7.790 sec)
INFO:tensorflow:lr = 0.0002567349 (7.791 sec)
INFO:tensorflow:global_step/sec: 12.6823
INFO:tensorflow:loss = 1.2701207, step = 44220 (7.886 sec)
INFO:tensorflow:lr = 0.000256477 (7.886 sec)
INFO:tensorflow:global_step/sec: 12.8214
INFO:tensorflow:loss = 1.2705317, step = 44320 (7.802 sec)
INFO:tensorflow:lr = 0.00025621938 (7.799 sec)
INFO:tensorflow:global_step/sec: 12.4534
INFO:tensorflow:loss = 1.2627566, step = 44420 (8.030 sec)
INFO:tensorflow:lr = 0.00025596202 (8.030 sec)
INFO:tensorflow:global_step/sec: 12.6966
INFO:tensorflow:loss = 1.2665414, step = 44520 (7.870 sec)
INFO:tensorflow:lr = 0.0002557049 (7.870 sec)
INFO:tensorflow:global_step/sec: 12.9501
INFO:tensorflow:loss = 1.2893751, step = 44620 (7.722 sec)
INFO:tensorflow:lr = 0.00025544802 (7.722 sec)
INFO:tensorflow:global_step/sec: 13.0372
INFO:tensorflow:loss = 1.2678509, step = 44720 (7.670 sec)
INFO:tensorflow:lr = 0.0002551914 (7.670 sec)
INFO:tensorflow:global_step/sec: 12.8701
INFO:tensorflow:loss = 1.2665724, step = 44820 (7.770 sec)
INFO:tensorflow:lr = 0.00025493503 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.7671
INFO:tensorflow:loss = 1.266214, step = 44920 (7.839 sec)
INFO:tensorflow:lr = 0.00025467898 (7.838 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.40537
INFO:tensorflow:loss = 1.2608075, step = 45020 (10.633 sec)
INFO:tensorflow:lr = 0.00025442315 (10.633 sec)
INFO:tensorflow:global_step/sec: 12.7037
INFO:tensorflow:loss = 1.264033, step = 45120 (7.869 sec)
INFO:tensorflow:lr = 0.00025416756 (7.869 sec)
INFO:tensorflow:global_step/sec: 12.8665
INFO:tensorflow:loss = 1.2578142, step = 45220 (7.774 sec)
INFO:tensorflow:lr = 0.00025391224 (7.774 sec)
INFO:tensorflow:global_step/sec: 12.4913
INFO:tensorflow:loss = 1.2547925, step = 45320 (8.004 sec)
INFO:tensorflow:lr = 0.0002536572 (8.004 sec)
INFO:tensorflow:global_step/sec: 12.595
INFO:tensorflow:loss = 1.2743986, step = 45420 (7.937 sec)
INFO:tensorflow:lr = 0.0002534024 (7.939 sec)
INFO:tensorflow:global_step/sec: 12.6593
INFO:tensorflow:loss = 1.266757, step = 45520 (7.897 sec)
INFO:tensorflow:lr = 0.00025314782 (7.899 sec)
INFO:tensorflow:global_step/sec: 12.8904
INFO:tensorflow:loss = 1.2690803, step = 45620 (7.760 sec)
INFO:tensorflow:lr = 0.00025289354 (7.759 sec)
INFO:tensorflow:global_step/sec: 12.8386
INFO:tensorflow:loss = 1.2695453, step = 45720 (7.792 sec)
INFO:tensorflow:lr = 0.0002526395 (7.790 sec)
INFO:tensorflow:global_step/sec: 12.6332
INFO:tensorflow:loss = 1.2680589, step = 45820 (7.911 sec)
INFO:tensorflow:lr = 0.0002523857 (7.910 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.4824
INFO:tensorflow:loss = 1.2544335, step = 45920 (8.014 sec)
INFO:tensorflow:lr = 0.0002521322 (8.014 sec)
INFO:tensorflow:global_step/sec: 10.8606
INFO:tensorflow:loss = 1.261169, step = 46020 (9.218 sec)
INFO:tensorflow:lr = 0.00025187893 (9.220 sec)
INFO:tensorflow:global_step/sec: 12.6674
INFO:tensorflow:loss = 1.2620407, step = 46120 (7.885 sec)
INFO:tensorflow:lr = 0.0002516259 (7.884 sec)
INFO:tensorflow:global_step/sec: 12.6643
INFO:tensorflow:loss = 1.2677541, step = 46220 (7.893 sec)
INFO:tensorflow:lr = 0.00025137313 (7.893 sec)
INFO:tensorflow:global_step/sec: 12.7899
INFO:tensorflow:loss = 1.254288, step = 46320 (7.818 sec)
INFO:tensorflow:lr = 0.00025112063 (7.819 sec)
INFO:tensorflow:global_step/sec: 12.7355
INFO:tensorflow:loss = 1.2622324, step = 46420 (7.851 sec)
INFO:tensorflow:lr = 0.00025086835 (7.850 sec)
INFO:tensorflow:global_step/sec: 12.77
INFO:tensorflow:loss = 1.2603494, step = 46520 (7.838 sec)
INFO:tensorflow:lr = 0.00025061634 (7.838 sec)
INFO:tensorflow:global_step/sec: 12.8105
INFO:tensorflow:loss = 1.274346, step = 46620 (7.801 sec)
INFO:tensorflow:lr = 0.0002503646 (7.801 sec)
INFO:tensorflow:global_step/sec: 12.8057
INFO:tensorflow:loss = 1.2791383, step = 46720 (7.808 sec)
INFO:tensorflow:lr = 0.0002501131 (7.809 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.6813
INFO:tensorflow:loss = 1.2684876, step = 46820 (7.892 sec)
INFO:tensorflow:lr = 0.00024986186 (7.890 sec)
INFO:tensorflow:global_step/sec: 11.2116
INFO:tensorflow:loss = 1.261704, step = 46920 (8.916 sec)
INFO:tensorflow:lr = 0.00024961086 (8.916 sec)
INFO:tensorflow:global_step/sec: 12.5528
INFO:tensorflow:loss = 1.2601739, step = 47020 (7.968 sec)
INFO:tensorflow:lr = 0.00024936014 (7.969 sec)
INFO:tensorflow:global_step/sec: 12.4465
INFO:tensorflow:loss = 1.2588917, step = 47120 (8.035 sec)
INFO:tensorflow:lr = 0.0002491096 (8.035 sec)
INFO:tensorflow:global_step/sec: 12.875
INFO:tensorflow:loss = 1.271802, step = 47220 (7.764 sec)
INFO:tensorflow:lr = 0.0002488594 (7.765 sec)
INFO:tensorflow:global_step/sec: 12.8335
INFO:tensorflow:loss = 1.263241, step = 47320 (7.789 sec)
INFO:tensorflow:lr = 0.00024860943 (7.788 sec)
INFO:tensorflow:global_step/sec: 13.0207
INFO:tensorflow:loss = 1.2612357, step = 47420 (7.685 sec)
INFO:tensorflow:lr = 0.0002483597 (7.685 sec)
INFO:tensorflow:global_step/sec: 12.6693
INFO:tensorflow:loss = 1.2592645, step = 47520 (7.888 sec)
INFO:tensorflow:lr = 0.0002481102 (7.890 sec)
INFO:tensorflow:global_step/sec: 12.7021
INFO:tensorflow:loss = 1.2634231, step = 47620 (7.877 sec)
INFO:tensorflow:lr = 0.00024786094 (7.876 sec)
INFO:tensorflow:global_step/sec: 12.7586
INFO:tensorflow:loss = 1.2658721, step = 47720 (7.837 sec)
INFO:tensorflow:lr = 0.00024761196 (7.842 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.901
INFO:tensorflow:loss = 1.2613206, step = 47820 (8.403 sec)
INFO:tensorflow:lr = 0.00024736323 (8.400 sec)
INFO:tensorflow:global_step/sec: 11.2274
INFO:tensorflow:loss = 1.2668675, step = 47920 (8.908 sec)
INFO:tensorflow:lr = 0.00024711475 (8.907 sec)
INFO:tensorflow:global_step/sec: 12.6344
INFO:tensorflow:loss = 1.2707262, step = 48020 (7.912 sec)
INFO:tensorflow:lr = 0.00024686652 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.6208
INFO:tensorflow:loss = 1.264168, step = 48120 (7.926 sec)
INFO:tensorflow:lr = 0.00024661855 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.4144
INFO:tensorflow:loss = 1.2581247, step = 48220 (8.055 sec)
INFO:tensorflow:lr = 0.00024637082 (8.055 sec)
INFO:tensorflow:global_step/sec: 12.1425
INFO:tensorflow:loss = 1.2826554, step = 48320 (8.229 sec)
INFO:tensorflow:lr = 0.00024612332 (8.231 sec)
INFO:tensorflow:global_step/sec: 12.3167
INFO:tensorflow:loss = 1.2596526, step = 48420 (8.125 sec)
INFO:tensorflow:lr = 0.0002458761 (8.124 sec)
INFO:tensorflow:global_step/sec: 12.5863
INFO:tensorflow:loss = 1.2663782, step = 48520 (7.940 sec)
INFO:tensorflow:lr = 0.00024562908 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.4722
INFO:tensorflow:loss = 1.2596316, step = 48620 (8.022 sec)
INFO:tensorflow:lr = 0.00024538234 (8.019 sec)
INFO:tensorflow:global_step/sec: 12.3972
INFO:tensorflow:loss = 1.2628276, step = 48720 (8.062 sec)
INFO:tensorflow:lr = 0.00024513586 (8.062 sec)
INFO:tensorflow:global_step/sec: 12.812
INFO:tensorflow:loss = 1.267219, step = 48820 (7.805 sec)
INFO:tensorflow:lr = 0.0002448896 (7.805 sec)
INFO:tensorflow:Saving checkpoints for 48900 into ../model/lstm_transformer/model.ckpt.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/saver.py:963: remove_checkpoint (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.
Instructions for updating:
Use standard file APIs to delete files with this prefix.
INFO:tensorflow:Loss for final step: 1.2682308.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-48900
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-48900
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.706
INFO:tensorflow:Best (Exact Match) Accuracy: 0.707
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-48900
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 48900 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.263123, step = 48900
INFO:tensorflow:lr = 0.00024469278
INFO:tensorflow:global_step/sec: 9.80343
INFO:tensorflow:loss = 1.2753992, step = 49000 (10.205 sec)
INFO:tensorflow:lr = 0.000244447 (10.206 sec)
INFO:tensorflow:global_step/sec: 12.3248
INFO:tensorflow:loss = 1.2618505, step = 49100 (8.117 sec)
INFO:tensorflow:lr = 0.00024420145 (8.114 sec)
INFO:tensorflow:global_step/sec: 12.7277
INFO:tensorflow:loss = 1.2551093, step = 49200 (7.851 sec)
INFO:tensorflow:lr = 0.00024395615 (7.852 sec)
INFO:tensorflow:global_step/sec: 13.0713
INFO:tensorflow:loss = 1.2688679, step = 49300 (7.651 sec)
INFO:tensorflow:lr = 0.00024371107 (7.650 sec)
INFO:tensorflow:global_step/sec: 12.8647
INFO:tensorflow:loss = 1.2720437, step = 49400 (7.774 sec)
INFO:tensorflow:lr = 0.00024346625 (7.775 sec)
INFO:tensorflow:global_step/sec: 12.7116
INFO:tensorflow:loss = 1.2706873, step = 49500 (7.866 sec)
INFO:tensorflow:lr = 0.00024322169 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.5179
INFO:tensorflow:loss = 1.2638872, step = 49600 (7.992 sec)
INFO:tensorflow:lr = 0.00024297737 (7.991 sec)
INFO:tensorflow:global_step/sec: 12.5553
INFO:tensorflow:loss = 1.2629794, step = 49700 (7.961 sec)
INFO:tensorflow:lr = 0.00024273328 (7.961 sec)
INFO:tensorflow:global_step/sec: 12.9475
INFO:tensorflow:loss = 1.2555556, step = 49800 (7.729 sec)
INFO:tensorflow:lr = 0.00024248945 (7.728 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.31103
INFO:tensorflow:loss = 1.2618929, step = 49900 (12.026 sec)
INFO:tensorflow:lr = 0.00024224588 (12.026 sec)
INFO:tensorflow:global_step/sec: 12.9152
INFO:tensorflow:loss = 1.2585716, step = 50000 (7.749 sec)
INFO:tensorflow:lr = 0.00024200253 (7.749 sec)
INFO:tensorflow:global_step/sec: 12.8223
INFO:tensorflow:loss = 1.2602012, step = 50100 (7.793 sec)
INFO:tensorflow:lr = 0.00024175941 (7.794 sec)
INFO:tensorflow:global_step/sec: 13.0814
INFO:tensorflow:loss = 1.2600421, step = 50200 (7.650 sec)
INFO:tensorflow:lr = 0.00024151658 (7.649 sec)
INFO:tensorflow:global_step/sec: 12.9555
INFO:tensorflow:loss = 1.2640244, step = 50300 (7.713 sec)
INFO:tensorflow:lr = 0.00024127394 (7.713 sec)
INFO:tensorflow:global_step/sec: 12.7933
INFO:tensorflow:loss = 1.2591022, step = 50400 (7.820 sec)
INFO:tensorflow:lr = 0.0002410316 (7.825 sec)
INFO:tensorflow:global_step/sec: 12.3763
INFO:tensorflow:loss = 1.2563342, step = 50500 (8.080 sec)
INFO:tensorflow:lr = 0.00024078948 (8.076 sec)
INFO:tensorflow:global_step/sec: 12.6579
INFO:tensorflow:loss = 1.251954, step = 50600 (7.897 sec)
INFO:tensorflow:lr = 0.0002405476 (7.896 sec)
INFO:tensorflow:global_step/sec: 12.8405
INFO:tensorflow:loss = 1.2703358, step = 50700 (7.795 sec)
INFO:tensorflow:lr = 0.00024030596 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.7567
INFO:tensorflow:loss = 1.260641, step = 50800 (7.839 sec)
INFO:tensorflow:lr = 0.00024006456 (7.839 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.37615
INFO:tensorflow:loss = 1.2584833, step = 50900 (11.932 sec)
INFO:tensorflow:lr = 0.00023982344 (11.932 sec)
INFO:tensorflow:global_step/sec: 12.8102
INFO:tensorflow:loss = 1.2630175, step = 51000 (7.805 sec)
INFO:tensorflow:lr = 0.00023958248 (7.805 sec)
INFO:tensorflow:global_step/sec: 12.8743
INFO:tensorflow:loss = 1.2574093, step = 51100 (7.768 sec)
INFO:tensorflow:lr = 0.00023934185 (7.768 sec)
INFO:tensorflow:global_step/sec: 12.9835
INFO:tensorflow:loss = 1.2609439, step = 51200 (7.705 sec)
INFO:tensorflow:lr = 0.0002391014 (7.705 sec)
INFO:tensorflow:global_step/sec: 12.8867
INFO:tensorflow:loss = 1.2654866, step = 51300 (7.757 sec)
INFO:tensorflow:lr = 0.00023886122 (7.756 sec)
INFO:tensorflow:global_step/sec: 12.7474
INFO:tensorflow:loss = 1.2698414, step = 51400 (7.844 sec)
INFO:tensorflow:lr = 0.00023862127 (7.844 sec)
INFO:tensorflow:global_step/sec: 12.7932
INFO:tensorflow:loss = 1.2593272, step = 51500 (7.817 sec)
INFO:tensorflow:lr = 0.00023838157 (7.821 sec)
INFO:tensorflow:global_step/sec: 12.6452
INFO:tensorflow:loss = 1.253849, step = 51600 (7.914 sec)
INFO:tensorflow:lr = 0.00023814212 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.6491
INFO:tensorflow:loss = 1.2664359, step = 51700 (7.906 sec)
INFO:tensorflow:lr = 0.00023790292 (7.905 sec)
INFO:tensorflow:global_step/sec: 12.6054
INFO:tensorflow:loss = 1.2593892, step = 51800 (7.927 sec)
INFO:tensorflow:lr = 0.0002376639 (7.928 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.3781
INFO:tensorflow:loss = 1.2685262, step = 51900 (11.940 sec)
INFO:tensorflow:lr = 0.0002374252 (11.941 sec)
INFO:tensorflow:global_step/sec: 12.5037
INFO:tensorflow:loss = 1.263798, step = 52000 (7.997 sec)
INFO:tensorflow:lr = 0.00023718669 (7.997 sec)
INFO:tensorflow:global_step/sec: 12.886
INFO:tensorflow:loss = 1.2689533, step = 52100 (7.756 sec)
INFO:tensorflow:lr = 0.00023694841 (7.756 sec)
INFO:tensorflow:global_step/sec: 12.7961
INFO:tensorflow:loss = 1.27047, step = 52200 (7.823 sec)
INFO:tensorflow:lr = 0.0002367104 (7.823 sec)
INFO:tensorflow:global_step/sec: 13.0623
INFO:tensorflow:loss = 1.2621255, step = 52300 (7.648 sec)
INFO:tensorflow:lr = 0.00023647262 (7.650 sec)
INFO:tensorflow:global_step/sec: 12.8008
INFO:tensorflow:loss = 1.270298, step = 52400 (7.812 sec)
INFO:tensorflow:lr = 0.00023623506 (7.811 sec)
INFO:tensorflow:global_step/sec: 12.8887
INFO:tensorflow:loss = 1.2604018, step = 52500 (7.760 sec)
INFO:tensorflow:lr = 0.00023599774 (7.761 sec)
INFO:tensorflow:global_step/sec: 12.717
INFO:tensorflow:loss = 1.2635633, step = 52600 (7.862 sec)
INFO:tensorflow:lr = 0.0002357607 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.6579
INFO:tensorflow:loss = 1.2612284, step = 52700 (7.900 sec)
INFO:tensorflow:lr = 0.00023552388 (7.897 sec)
INFO:tensorflow:global_step/sec: 12.7953
INFO:tensorflow:loss = 1.258312, step = 52800 (7.821 sec)
INFO:tensorflow:lr = 0.0002352873 (7.820 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.38183
INFO:tensorflow:loss = 1.2634532, step = 52900 (11.925 sec)
INFO:tensorflow:lr = 0.00023505092 (11.929 sec)
INFO:tensorflow:global_step/sec: 12.5748
INFO:tensorflow:loss = 1.2694097, step = 53000 (7.953 sec)
INFO:tensorflow:lr = 0.00023481481 (7.949 sec)
INFO:tensorflow:global_step/sec: 12.8424
INFO:tensorflow:loss = 1.2625628, step = 53100 (7.792 sec)
INFO:tensorflow:lr = 0.00023457894 (7.791 sec)
INFO:tensorflow:global_step/sec: 12.5576
INFO:tensorflow:loss = 1.260535, step = 53200 (7.964 sec)
INFO:tensorflow:lr = 0.00023434329 (7.965 sec)
INFO:tensorflow:global_step/sec: 12.7491
INFO:tensorflow:loss = 1.2549899, step = 53300 (7.843 sec)
INFO:tensorflow:lr = 0.0002341079 (7.843 sec)
INFO:tensorflow:global_step/sec: 12.7266
INFO:tensorflow:loss = 1.2714366, step = 53400 (7.852 sec)
INFO:tensorflow:lr = 0.00023387272 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.7257
INFO:tensorflow:loss = 1.2598583, step = 53500 (7.863 sec)
INFO:tensorflow:lr = 0.00023363781 (7.853 sec)
INFO:tensorflow:global_step/sec: 12.9742
INFO:tensorflow:loss = 1.2618462, step = 53600 (7.710 sec)
INFO:tensorflow:lr = 0.0002334031 (7.707 sec)
INFO:tensorflow:global_step/sec: 12.7653
INFO:tensorflow:loss = 1.2702688, step = 53700 (7.828 sec)
INFO:tensorflow:lr = 0.00023316864 (7.828 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.79073
INFO:tensorflow:loss = 1.2538612, step = 53800 (11.378 sec)
INFO:tensorflow:lr = 0.00023293443 (11.380 sec)
INFO:tensorflow:global_step/sec: 13.0106
INFO:tensorflow:loss = 1.2529118, step = 53900 (7.687 sec)
INFO:tensorflow:lr = 0.00023270043 (7.685 sec)
INFO:tensorflow:global_step/sec: 12.9886
INFO:tensorflow:loss = 1.2524719, step = 54000 (7.694 sec)
INFO:tensorflow:lr = 0.00023246667 (7.695 sec)
INFO:tensorflow:global_step/sec: 12.6384
INFO:tensorflow:loss = 1.2778753, step = 54100 (7.917 sec)
INFO:tensorflow:lr = 0.00023223316 (7.917 sec)
INFO:tensorflow:global_step/sec: 12.5824
INFO:tensorflow:loss = 1.2629167, step = 54200 (7.943 sec)
INFO:tensorflow:lr = 0.00023199987 (7.946 sec)
INFO:tensorflow:global_step/sec: 12.8335
INFO:tensorflow:loss = 1.2570986, step = 54300 (7.792 sec)
INFO:tensorflow:lr = 0.00023176681 (7.789 sec)
INFO:tensorflow:global_step/sec: 12.6558
INFO:tensorflow:loss = 1.2723862, step = 54400 (7.902 sec)
INFO:tensorflow:lr = 0.00023153402 (7.902 sec)
INFO:tensorflow:global_step/sec: 12.8133
INFO:tensorflow:loss = 1.256413, step = 54500 (7.804 sec)
INFO:tensorflow:lr = 0.00023130143 (7.803 sec)
INFO:tensorflow:global_step/sec: 12.7269
INFO:tensorflow:loss = 1.2575724, step = 54600 (7.857 sec)
INFO:tensorflow:lr = 0.00023106908 (7.858 sec)
INFO:tensorflow:global_step/sec: 12.5403
INFO:tensorflow:loss = 1.272584, step = 54700 (7.975 sec)
INFO:tensorflow:lr = 0.00023083696 (7.976 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.19518
INFO:tensorflow:loss = 1.2527137, step = 54800 (10.875 sec)
INFO:tensorflow:lr = 0.0002306051 (10.876 sec)
INFO:tensorflow:global_step/sec: 12.491
INFO:tensorflow:loss = 1.2544173, step = 54900 (8.012 sec)
INFO:tensorflow:lr = 0.00023037342 (8.009 sec)
INFO:tensorflow:global_step/sec: 13.0044
INFO:tensorflow:loss = 1.2695364, step = 55000 (7.683 sec)
INFO:tensorflow:lr = 0.00023014202 (7.691 sec)
INFO:tensorflow:global_step/sec: 12.6771
INFO:tensorflow:loss = 1.2688967, step = 55100 (7.888 sec)
INFO:tensorflow:lr = 0.00022991082 (7.880 sec)
INFO:tensorflow:global_step/sec: 12.7627
INFO:tensorflow:loss = 1.2663212, step = 55200 (7.835 sec)
INFO:tensorflow:lr = 0.00022967989 (7.839 sec)
INFO:tensorflow:global_step/sec: 12.6703
INFO:tensorflow:loss = 1.2508467, step = 55300 (7.897 sec)
INFO:tensorflow:lr = 0.00022944914 (7.894 sec)
INFO:tensorflow:global_step/sec: 12.7355
INFO:tensorflow:loss = 1.2672036, step = 55400 (7.849 sec)
INFO:tensorflow:lr = 0.00022921867 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.8933
INFO:tensorflow:loss = 1.2606906, step = 55500 (7.755 sec)
INFO:tensorflow:lr = 0.0002289884 (7.759 sec)
INFO:tensorflow:global_step/sec: 12.7887
INFO:tensorflow:loss = 1.2555453, step = 55600 (7.826 sec)
INFO:tensorflow:lr = 0.0002287584 (7.825 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.3639
INFO:tensorflow:loss = 1.2551997, step = 55700 (8.091 sec)
INFO:tensorflow:lr = 0.00022852859 (8.089 sec)
INFO:tensorflow:global_step/sec: 10.6991
INFO:tensorflow:loss = 1.2652216, step = 55800 (9.337 sec)
INFO:tensorflow:lr = 0.00022829902 (9.342 sec)
INFO:tensorflow:global_step/sec: 12.824
INFO:tensorflow:loss = 1.2569361, step = 55900 (7.800 sec)
INFO:tensorflow:lr = 0.00022806968 (7.797 sec)
INFO:tensorflow:global_step/sec: 12.794
INFO:tensorflow:loss = 1.2683395, step = 56000 (7.815 sec)
INFO:tensorflow:lr = 0.00022784059 (7.812 sec)
INFO:tensorflow:global_step/sec: 12.9955
INFO:tensorflow:loss = 1.264496, step = 56100 (7.695 sec)
INFO:tensorflow:lr = 0.00022761173 (7.695 sec)
INFO:tensorflow:global_step/sec: 13.0333
INFO:tensorflow:loss = 1.2644389, step = 56200 (7.672 sec)
INFO:tensorflow:lr = 0.00022738309 (7.676 sec)
INFO:tensorflow:global_step/sec: 13.022
INFO:tensorflow:loss = 1.253439, step = 56300 (7.679 sec)
INFO:tensorflow:lr = 0.00022715465 (7.676 sec)
INFO:tensorflow:global_step/sec: 12.5266
INFO:tensorflow:loss = 1.2558391, step = 56400 (7.983 sec)
INFO:tensorflow:lr = 0.00022692647 (7.984 sec)
INFO:tensorflow:global_step/sec: 12.7366
INFO:tensorflow:loss = 1.2584279, step = 56500 (7.855 sec)
INFO:tensorflow:lr = 0.00022669851 (7.854 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2942
INFO:tensorflow:loss = 1.2605096, step = 56600 (8.132 sec)
INFO:tensorflow:lr = 0.00022647077 (8.133 sec)
INFO:tensorflow:global_step/sec: 10.8588
INFO:tensorflow:loss = 1.2645557, step = 56700 (9.207 sec)
INFO:tensorflow:lr = 0.0002262433 (9.205 sec)
INFO:tensorflow:global_step/sec: 12.9058
INFO:tensorflow:loss = 1.257235, step = 56800 (7.751 sec)
INFO:tensorflow:lr = 0.00022601604 (7.751 sec)
INFO:tensorflow:global_step/sec: 12.6226
INFO:tensorflow:loss = 1.259087, step = 56900 (7.924 sec)
INFO:tensorflow:lr = 0.00022578899 (7.925 sec)
INFO:tensorflow:global_step/sec: 12.6459
INFO:tensorflow:loss = 1.2512816, step = 57000 (7.905 sec)
INFO:tensorflow:lr = 0.00022556218 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.8169
INFO:tensorflow:loss = 1.256032, step = 57100 (7.800 sec)
INFO:tensorflow:lr = 0.00022533562 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.7711
INFO:tensorflow:loss = 1.2535781, step = 57200 (7.830 sec)
INFO:tensorflow:lr = 0.00022510927 (7.829 sec)
INFO:tensorflow:global_step/sec: 12.7293
INFO:tensorflow:loss = 1.257878, step = 57300 (7.863 sec)
INFO:tensorflow:lr = 0.0002248831 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.5132
INFO:tensorflow:loss = 1.2791443, step = 57400 (7.988 sec)
INFO:tensorflow:lr = 0.00022465723 (7.987 sec)
INFO:tensorflow:global_step/sec: 13.0392
INFO:tensorflow:loss = 1.2567543, step = 57500 (7.668 sec)
INFO:tensorflow:lr = 0.00022443154 (7.669 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1479
INFO:tensorflow:loss = 1.2558892, step = 57600 (8.235 sec)
INFO:tensorflow:lr = 0.0002242061 (8.235 sec)
INFO:tensorflow:global_step/sec: 11.7853
INFO:tensorflow:loss = 1.270159, step = 57700 (8.479 sec)
INFO:tensorflow:lr = 0.00022398087 (8.480 sec)
INFO:tensorflow:global_step/sec: 12.9312
INFO:tensorflow:loss = 1.2561911, step = 57800 (7.738 sec)
INFO:tensorflow:lr = 0.00022375588 (7.737 sec)
INFO:tensorflow:global_step/sec: 12.8943
INFO:tensorflow:loss = 1.2581128, step = 57900 (7.751 sec)
INFO:tensorflow:lr = 0.0002235311 (7.751 sec)
INFO:tensorflow:global_step/sec: 12.89
INFO:tensorflow:loss = 1.2589202, step = 58000 (7.763 sec)
INFO:tensorflow:lr = 0.00022330656 (7.763 sec)
INFO:tensorflow:global_step/sec: 13.0328
INFO:tensorflow:loss = 1.2635732, step = 58100 (7.668 sec)
INFO:tensorflow:lr = 0.00022308226 (7.670 sec)
INFO:tensorflow:global_step/sec: 12.9888
INFO:tensorflow:loss = 1.2706214, step = 58200 (7.699 sec)
INFO:tensorflow:lr = 0.00022285814 (7.699 sec)
INFO:tensorflow:global_step/sec: 12.7051
INFO:tensorflow:loss = 1.258386, step = 58300 (7.871 sec)
INFO:tensorflow:lr = 0.00022263429 (7.879 sec)
INFO:tensorflow:global_step/sec: 13.0078
INFO:tensorflow:loss = 1.2566429, step = 58400 (7.687 sec)
INFO:tensorflow:lr = 0.00022241064 (7.676 sec)
INFO:tensorflow:global_step/sec: 12.6443
INFO:tensorflow:loss = 1.2667817, step = 58500 (7.910 sec)
INFO:tensorflow:lr = 0.00022218723 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.7619
INFO:tensorflow:loss = 1.2693475, step = 58600 (7.835 sec)
INFO:tensorflow:lr = 0.00022196402 (7.835 sec)
INFO:tensorflow:Saving checkpoints for 58680 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2575451.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-58680
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-58680
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.714
INFO:tensorflow:Best (Exact Match) Accuracy: 0.714
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-58680
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 58680 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2724138, step = 58680
INFO:tensorflow:lr = 0.00022178564
INFO:tensorflow:global_step/sec: 10.3144
INFO:tensorflow:loss = 1.2573106, step = 58780 (9.698 sec)
INFO:tensorflow:lr = 0.00022156286 (9.706 sec)
INFO:tensorflow:global_step/sec: 12.8129
INFO:tensorflow:loss = 1.2532308, step = 58880 (7.803 sec)
INFO:tensorflow:lr = 0.00022134029 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.6229
INFO:tensorflow:loss = 1.2531558, step = 58980 (7.928 sec)
INFO:tensorflow:lr = 0.00022111794 (7.928 sec)
INFO:tensorflow:global_step/sec: 12.8638
INFO:tensorflow:loss = 1.2645673, step = 59080 (7.773 sec)
INFO:tensorflow:lr = 0.0002208958 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.4406
INFO:tensorflow:loss = 1.2638228, step = 59180 (8.039 sec)
INFO:tensorflow:lr = 0.00022067393 (8.038 sec)
INFO:tensorflow:global_step/sec: 12.6973
INFO:tensorflow:loss = 1.2559628, step = 59280 (7.877 sec)
INFO:tensorflow:lr = 0.00022045226 (7.877 sec)
INFO:tensorflow:global_step/sec: 12.6856
INFO:tensorflow:loss = 1.256625, step = 59380 (7.876 sec)
INFO:tensorflow:lr = 0.00022023081 (7.877 sec)
INFO:tensorflow:global_step/sec: 13.3057
INFO:tensorflow:loss = 1.2582057, step = 59480 (7.517 sec)
INFO:tensorflow:lr = 0.00022000956 (7.517 sec)
INFO:tensorflow:global_step/sec: 12.5282
INFO:tensorflow:loss = 1.2669188, step = 59580 (7.982 sec)
INFO:tensorflow:lr = 0.00021978856 (7.986 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.22968
INFO:tensorflow:loss = 1.2529101, step = 59680 (12.156 sec)
INFO:tensorflow:lr = 0.00021956777 (12.151 sec)
INFO:tensorflow:global_step/sec: 12.8956
INFO:tensorflow:loss = 1.2675282, step = 59780 (7.755 sec)
INFO:tensorflow:lr = 0.00021934723 (7.760 sec)
INFO:tensorflow:global_step/sec: 12.6672
INFO:tensorflow:loss = 1.2623235, step = 59880 (7.891 sec)
INFO:tensorflow:lr = 0.00021912686 (7.889 sec)
INFO:tensorflow:global_step/sec: 12.5604
INFO:tensorflow:loss = 1.2686803, step = 59980 (7.958 sec)
INFO:tensorflow:lr = 0.00021890676 (7.956 sec)
INFO:tensorflow:global_step/sec: 12.9026
INFO:tensorflow:loss = 1.2615492, step = 60080 (7.751 sec)
INFO:tensorflow:lr = 0.00021868687 (7.753 sec)
INFO:tensorflow:global_step/sec: 12.8044
INFO:tensorflow:loss = 1.2547079, step = 60180 (7.809 sec)
INFO:tensorflow:lr = 0.00021846718 (7.807 sec)
INFO:tensorflow:global_step/sec: 12.8346
INFO:tensorflow:loss = 1.2528274, step = 60280 (7.792 sec)
INFO:tensorflow:lr = 0.00021824775 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.9208
INFO:tensorflow:loss = 1.2669843, step = 60380 (7.741 sec)
INFO:tensorflow:lr = 0.0002180285 (7.738 sec)
INFO:tensorflow:global_step/sec: 12.735
INFO:tensorflow:loss = 1.2558151, step = 60480 (7.850 sec)
INFO:tensorflow:lr = 0.00021780949 (7.849 sec)
INFO:tensorflow:global_step/sec: 12.8776
INFO:tensorflow:loss = 1.2551497, step = 60580 (7.771 sec)
INFO:tensorflow:lr = 0.00021759068 (7.771 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.34405
INFO:tensorflow:loss = 1.2546343, step = 60680 (11.980 sec)
INFO:tensorflow:lr = 0.0002173721 (11.980 sec)
INFO:tensorflow:global_step/sec: 12.7156
INFO:tensorflow:loss = 1.2520571, step = 60780 (7.867 sec)
INFO:tensorflow:lr = 0.00021715376 (7.867 sec)
INFO:tensorflow:global_step/sec: 12.6516
INFO:tensorflow:loss = 1.2656932, step = 60880 (7.904 sec)
INFO:tensorflow:lr = 0.00021693563 (7.905 sec)
INFO:tensorflow:global_step/sec: 13.0749
INFO:tensorflow:loss = 1.2510619, step = 60980 (7.650 sec)
INFO:tensorflow:lr = 0.00021671772 (7.650 sec)
INFO:tensorflow:global_step/sec: 12.6946
INFO:tensorflow:loss = 1.2555563, step = 61080 (7.872 sec)
INFO:tensorflow:lr = 0.00021650002 (7.872 sec)
INFO:tensorflow:global_step/sec: 12.8617
INFO:tensorflow:loss = 1.2622253, step = 61180 (7.778 sec)
INFO:tensorflow:lr = 0.00021628253 (7.778 sec)
INFO:tensorflow:global_step/sec: 12.7325
INFO:tensorflow:loss = 1.2601289, step = 61280 (7.855 sec)
INFO:tensorflow:lr = 0.00021606525 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.7575
INFO:tensorflow:loss = 1.2692169, step = 61380 (7.834 sec)
INFO:tensorflow:lr = 0.00021584822 (7.833 sec)
INFO:tensorflow:global_step/sec: 12.6755
INFO:tensorflow:loss = 1.2582502, step = 61480 (7.889 sec)
INFO:tensorflow:lr = 0.00021563139 (7.890 sec)
INFO:tensorflow:global_step/sec: 12.9292
INFO:tensorflow:loss = 1.2521135, step = 61580 (7.739 sec)
INFO:tensorflow:lr = 0.00021541478 (7.739 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.3132
INFO:tensorflow:loss = 1.2625241, step = 61680 (12.031 sec)
INFO:tensorflow:lr = 0.0002151984 (12.030 sec)
INFO:tensorflow:global_step/sec: 12.8491
INFO:tensorflow:loss = 1.2569088, step = 61780 (7.777 sec)
INFO:tensorflow:lr = 0.00021498224 (7.778 sec)
INFO:tensorflow:global_step/sec: 12.6945
INFO:tensorflow:loss = 1.2559131, step = 61880 (7.882 sec)
INFO:tensorflow:lr = 0.00021476626 (7.883 sec)
INFO:tensorflow:global_step/sec: 12.8822
INFO:tensorflow:loss = 1.2696959, step = 61980 (7.759 sec)
INFO:tensorflow:lr = 0.00021455051 (7.759 sec)
INFO:tensorflow:global_step/sec: 12.8127
INFO:tensorflow:loss = 1.2561213, step = 62080 (7.803 sec)
INFO:tensorflow:lr = 0.00021433501 (7.803 sec)
INFO:tensorflow:global_step/sec: 12.8856
INFO:tensorflow:loss = 1.2621582, step = 62180 (7.766 sec)
INFO:tensorflow:lr = 0.00021411969 (7.766 sec)
INFO:tensorflow:global_step/sec: 12.5122
INFO:tensorflow:loss = 1.2536796, step = 62280 (7.991 sec)
INFO:tensorflow:lr = 0.00021390461 (7.990 sec)
INFO:tensorflow:global_step/sec: 12.3773
INFO:tensorflow:loss = 1.2566583, step = 62380 (8.081 sec)
INFO:tensorflow:lr = 0.00021368972 (8.081 sec)
INFO:tensorflow:global_step/sec: 12.6472
INFO:tensorflow:loss = 1.2600583, step = 62480 (7.900 sec)
INFO:tensorflow:lr = 0.00021347508 (7.904 sec)
INFO:tensorflow:global_step/sec: 12.6644
INFO:tensorflow:loss = 1.2559776, step = 62580 (7.902 sec)
INFO:tensorflow:lr = 0.00021326065 (7.898 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.48244
INFO:tensorflow:loss = 1.2560581, step = 62680 (11.790 sec)
INFO:tensorflow:lr = 0.0002130464 (11.791 sec)
INFO:tensorflow:global_step/sec: 12.8693
INFO:tensorflow:loss = 1.2563652, step = 62780 (7.768 sec)
INFO:tensorflow:lr = 0.0002128324 (7.767 sec)
INFO:tensorflow:global_step/sec: 12.7061
INFO:tensorflow:loss = 1.2647818, step = 62880 (7.869 sec)
INFO:tensorflow:lr = 0.0002126186 (7.869 sec)
INFO:tensorflow:global_step/sec: 12.5926
INFO:tensorflow:loss = 1.2658582, step = 62980 (7.939 sec)
INFO:tensorflow:lr = 0.00021240504 (7.938 sec)
INFO:tensorflow:global_step/sec: 12.8705
INFO:tensorflow:loss = 1.2523164, step = 63080 (7.769 sec)
INFO:tensorflow:lr = 0.00021219168 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.7683
INFO:tensorflow:loss = 1.2602874, step = 63180 (7.837 sec)
INFO:tensorflow:lr = 0.0002119785 (7.842 sec)
INFO:tensorflow:global_step/sec: 13.0614
INFO:tensorflow:loss = 1.2652452, step = 63280 (7.655 sec)
INFO:tensorflow:lr = 0.00021176557 (7.650 sec)
INFO:tensorflow:global_step/sec: 12.6689
INFO:tensorflow:loss = 1.250535, step = 63380 (7.889 sec)
INFO:tensorflow:lr = 0.00021155283 (7.890 sec)
INFO:tensorflow:global_step/sec: 12.726
INFO:tensorflow:loss = 1.2587569, step = 63480 (7.859 sec)
INFO:tensorflow:lr = 0.00021134033 (7.869 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.5915
INFO:tensorflow:loss = 1.2486932, step = 63580 (11.639 sec)
INFO:tensorflow:lr = 0.00021112803 (11.630 sec)
INFO:tensorflow:global_step/sec: 12.746
INFO:tensorflow:loss = 1.2581683, step = 63680 (7.845 sec)
INFO:tensorflow:lr = 0.00021091594 (7.853 sec)
INFO:tensorflow:global_step/sec: 12.9087
INFO:tensorflow:loss = 1.2698822, step = 63780 (7.752 sec)
INFO:tensorflow:lr = 0.00021070408 (7.744 sec)
INFO:tensorflow:global_step/sec: 12.7608
INFO:tensorflow:loss = 1.2671998, step = 63880 (7.835 sec)
INFO:tensorflow:lr = 0.00021049242 (7.835 sec)
INFO:tensorflow:global_step/sec: 12.9193
INFO:tensorflow:loss = 1.2540299, step = 63980 (7.742 sec)
INFO:tensorflow:lr = 0.00021028098 (7.742 sec)
INFO:tensorflow:global_step/sec: 12.7305
INFO:tensorflow:loss = 1.2617122, step = 64080 (7.856 sec)
INFO:tensorflow:lr = 0.00021006975 (7.856 sec)
INFO:tensorflow:global_step/sec: 12.5839
INFO:tensorflow:loss = 1.2595084, step = 64180 (7.945 sec)
INFO:tensorflow:lr = 0.00020985873 (7.946 sec)
INFO:tensorflow:global_step/sec: 13.0107
INFO:tensorflow:loss = 1.2606783, step = 64280 (7.681 sec)
INFO:tensorflow:lr = 0.00020964789 (7.681 sec)
INFO:tensorflow:global_step/sec: 12.6722
INFO:tensorflow:loss = 1.2504494, step = 64380 (7.898 sec)
INFO:tensorflow:lr = 0.00020943732 (7.898 sec)
INFO:tensorflow:global_step/sec: 12.6237
INFO:tensorflow:loss = 1.2601182, step = 64480 (7.921 sec)
INFO:tensorflow:lr = 0.00020922691 (7.921 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.1461
INFO:tensorflow:loss = 1.2577064, step = 64580 (10.934 sec)
INFO:tensorflow:lr = 0.00020901677 (10.934 sec)
INFO:tensorflow:global_step/sec: 12.7236
INFO:tensorflow:loss = 1.2530557, step = 64680 (7.857 sec)
INFO:tensorflow:lr = 0.0002088068 (7.858 sec)
INFO:tensorflow:global_step/sec: 13.0769
INFO:tensorflow:loss = 1.2567884, step = 64780 (7.651 sec)
INFO:tensorflow:lr = 0.00020859703 (7.655 sec)
INFO:tensorflow:global_step/sec: 12.684
INFO:tensorflow:loss = 1.251407, step = 64880 (7.882 sec)
INFO:tensorflow:lr = 0.00020838752 (7.877 sec)
INFO:tensorflow:global_step/sec: 12.8542
INFO:tensorflow:loss = 1.2556778, step = 64980 (7.781 sec)
INFO:tensorflow:lr = 0.00020817819 (7.781 sec)
INFO:tensorflow:global_step/sec: 12.7122
INFO:tensorflow:loss = 1.2546712, step = 65080 (7.860 sec)
INFO:tensorflow:lr = 0.00020796905 (7.860 sec)
INFO:tensorflow:global_step/sec: 12.9692
INFO:tensorflow:loss = 1.2524756, step = 65180 (7.715 sec)
INFO:tensorflow:lr = 0.00020776014 (7.715 sec)
INFO:tensorflow:global_step/sec: 13.0259
INFO:tensorflow:loss = 1.2570271, step = 65280 (7.677 sec)
INFO:tensorflow:lr = 0.00020755142 (7.680 sec)
INFO:tensorflow:global_step/sec: 12.7252
INFO:tensorflow:loss = 1.25036, step = 65380 (7.860 sec)
INFO:tensorflow:lr = 0.00020734294 (7.859 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1057
INFO:tensorflow:loss = 1.2546037, step = 65480 (8.258 sec)
INFO:tensorflow:lr = 0.00020713465 (8.258 sec)
INFO:tensorflow:global_step/sec: 10.5931
INFO:tensorflow:loss = 1.2538458, step = 65580 (9.437 sec)
INFO:tensorflow:lr = 0.00020692658 (9.435 sec)
INFO:tensorflow:global_step/sec: 12.8133
INFO:tensorflow:loss = 1.2516928, step = 65680 (7.804 sec)
INFO:tensorflow:lr = 0.00020671873 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.9175
INFO:tensorflow:loss = 1.2501926, step = 65780 (7.746 sec)
INFO:tensorflow:lr = 0.00020651106 (7.751 sec)
INFO:tensorflow:global_step/sec: 13.1593
INFO:tensorflow:loss = 1.2550164, step = 65880 (7.601 sec)
INFO:tensorflow:lr = 0.00020630364 (7.596 sec)
INFO:tensorflow:global_step/sec: 12.7002
INFO:tensorflow:loss = 1.2505937, step = 65980 (7.874 sec)
INFO:tensorflow:lr = 0.00020609639 (7.873 sec)
INFO:tensorflow:global_step/sec: 12.373
INFO:tensorflow:loss = 1.2635084, step = 66080 (8.078 sec)
INFO:tensorflow:lr = 0.00020588936 (8.079 sec)
INFO:tensorflow:global_step/sec: 12.8026
INFO:tensorflow:loss = 1.2599782, step = 66180 (7.815 sec)
INFO:tensorflow:lr = 0.00020568253 (7.814 sec)
INFO:tensorflow:global_step/sec: 12.5392
INFO:tensorflow:loss = 1.2670157, step = 66280 (7.970 sec)
INFO:tensorflow:lr = 0.00020547592 (7.970 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2205
INFO:tensorflow:loss = 1.2632979, step = 66380 (8.189 sec)
INFO:tensorflow:lr = 0.0002052695 (8.188 sec)
INFO:tensorflow:global_step/sec: 11.0385
INFO:tensorflow:loss = 1.269494, step = 66480 (9.056 sec)
INFO:tensorflow:lr = 0.00020506332 (9.058 sec)
INFO:tensorflow:global_step/sec: 12.8458
INFO:tensorflow:loss = 1.2560563, step = 66580 (7.781 sec)
INFO:tensorflow:lr = 0.00020485735 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.5578
INFO:tensorflow:loss = 1.2637135, step = 66680 (7.969 sec)
INFO:tensorflow:lr = 0.00020465154 (7.969 sec)
INFO:tensorflow:global_step/sec: 12.8306
INFO:tensorflow:loss = 1.2596672, step = 66780 (7.794 sec)
INFO:tensorflow:lr = 0.00020444595 (7.793 sec)
INFO:tensorflow:global_step/sec: 12.9435
INFO:tensorflow:loss = 1.253951, step = 66880 (7.723 sec)
INFO:tensorflow:lr = 0.00020424058 (7.723 sec)
INFO:tensorflow:global_step/sec: 12.5873
INFO:tensorflow:loss = 1.2510319, step = 66980 (7.948 sec)
INFO:tensorflow:lr = 0.00020403544 (7.948 sec)
INFO:tensorflow:global_step/sec: 12.9418
INFO:tensorflow:loss = 1.2582647, step = 67080 (7.728 sec)
INFO:tensorflow:lr = 0.00020383045 (7.728 sec)
INFO:tensorflow:global_step/sec: 12.9173
INFO:tensorflow:loss = 1.2596313, step = 67180 (7.739 sec)
INFO:tensorflow:lr = 0.00020362572 (7.739 sec)
INFO:tensorflow:global_step/sec: 12.9503
INFO:tensorflow:loss = 1.2502924, step = 67280 (7.722 sec)
INFO:tensorflow:lr = 0.00020342116 (7.722 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1161
INFO:tensorflow:loss = 1.2566557, step = 67380 (8.256 sec)
INFO:tensorflow:lr = 0.00020321681 (8.255 sec)
INFO:tensorflow:global_step/sec: 11.3069
INFO:tensorflow:loss = 1.2655523, step = 67480 (8.838 sec)
INFO:tensorflow:lr = 0.00020301268 (8.838 sec)
INFO:tensorflow:global_step/sec: 12.8634
INFO:tensorflow:loss = 1.2517667, step = 67580 (7.774 sec)
INFO:tensorflow:lr = 0.00020280876 (7.774 sec)
INFO:tensorflow:global_step/sec: 12.9586
INFO:tensorflow:loss = 1.2534577, step = 67680 (7.717 sec)
INFO:tensorflow:lr = 0.00020260502 (7.716 sec)
INFO:tensorflow:global_step/sec: 12.8067
INFO:tensorflow:loss = 1.256631, step = 67780 (7.808 sec)
INFO:tensorflow:lr = 0.00020240148 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.8087
INFO:tensorflow:loss = 1.2600688, step = 67880 (7.807 sec)
INFO:tensorflow:lr = 0.00020219819 (7.808 sec)
INFO:tensorflow:global_step/sec: 12.6442
INFO:tensorflow:loss = 1.2547219, step = 67980 (7.909 sec)
INFO:tensorflow:lr = 0.00020199506 (7.909 sec)
INFO:tensorflow:global_step/sec: 12.7405
INFO:tensorflow:loss = 1.2526922, step = 68080 (7.849 sec)
INFO:tensorflow:lr = 0.0002017922 (7.849 sec)
INFO:tensorflow:global_step/sec: 12.9993
INFO:tensorflow:loss = 1.2611421, step = 68180 (7.692 sec)
INFO:tensorflow:lr = 0.00020158946 (7.693 sec)
INFO:tensorflow:global_step/sec: 12.5706
INFO:tensorflow:loss = 1.2646718, step = 68280 (7.955 sec)
INFO:tensorflow:lr = 0.00020138695 (7.958 sec)
INFO:tensorflow:global_step/sec: 12.4633
INFO:tensorflow:loss = 1.2613611, step = 68380 (8.024 sec)
INFO:tensorflow:lr = 0.00020118465 (8.023 sec)
INFO:tensorflow:Saving checkpoints for 68460 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2606231.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-68460
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-68460
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.717
INFO:tensorflow:Best (Exact Match) Accuracy: 0.717
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-68460
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 68460 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.250214, step = 68460
INFO:tensorflow:lr = 0.00020102295
INFO:tensorflow:global_step/sec: 9.89178
INFO:tensorflow:loss = 1.2552146, step = 68560 (10.113 sec)
INFO:tensorflow:lr = 0.00020082104 (10.114 sec)
INFO:tensorflow:global_step/sec: 12.6776
INFO:tensorflow:loss = 1.2509973, step = 68660 (7.886 sec)
INFO:tensorflow:lr = 0.00020061928 (7.887 sec)
INFO:tensorflow:global_step/sec: 12.514
INFO:tensorflow:loss = 1.2553383, step = 68760 (7.996 sec)
INFO:tensorflow:lr = 0.00020041778 (7.995 sec)
INFO:tensorflow:global_step/sec: 12.7919
INFO:tensorflow:loss = 1.2552305, step = 68860 (7.817 sec)
INFO:tensorflow:lr = 0.00020021645 (7.818 sec)
INFO:tensorflow:global_step/sec: 12.8296
INFO:tensorflow:loss = 1.2542683, step = 68960 (7.791 sec)
INFO:tensorflow:lr = 0.0002000153 (7.790 sec)
INFO:tensorflow:global_step/sec: 12.5442
INFO:tensorflow:loss = 1.2532531, step = 69060 (7.971 sec)
INFO:tensorflow:lr = 0.0001998144 (7.970 sec)
INFO:tensorflow:global_step/sec: 12.781
INFO:tensorflow:loss = 1.2514119, step = 69160 (7.830 sec)
INFO:tensorflow:lr = 0.00019961369 (7.831 sec)
INFO:tensorflow:global_step/sec: 13.0693
INFO:tensorflow:loss = 1.2642561, step = 69260 (7.645 sec)
INFO:tensorflow:lr = 0.00019941316 (7.645 sec)
INFO:tensorflow:global_step/sec: 12.9636
INFO:tensorflow:loss = 1.2500038, step = 69360 (7.714 sec)
INFO:tensorflow:lr = 0.00019921285 (7.714 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.32059
INFO:tensorflow:loss = 1.2585869, step = 69460 (12.018 sec)
INFO:tensorflow:lr = 0.00019901272 (12.018 sec)
INFO:tensorflow:global_step/sec: 12.9882
INFO:tensorflow:loss = 1.2621295, step = 69560 (7.704 sec)
INFO:tensorflow:lr = 0.00019881282 (7.705 sec)
INFO:tensorflow:global_step/sec: 12.653
INFO:tensorflow:loss = 1.2571415, step = 69660 (7.898 sec)
INFO:tensorflow:lr = 0.0001986131 (7.902 sec)
INFO:tensorflow:global_step/sec: 12.3987
INFO:tensorflow:loss = 1.252396, step = 69760 (8.065 sec)
INFO:tensorflow:lr = 0.00019841359 (8.064 sec)
INFO:tensorflow:global_step/sec: 12.7307
INFO:tensorflow:loss = 1.2507378, step = 69860 (7.860 sec)
INFO:tensorflow:lr = 0.0001982143 (7.858 sec)
INFO:tensorflow:global_step/sec: 13.0388
INFO:tensorflow:loss = 1.254819, step = 69960 (7.666 sec)
INFO:tensorflow:lr = 0.00019801517 (7.665 sec)
INFO:tensorflow:global_step/sec: 12.7131
INFO:tensorflow:loss = 1.2549182, step = 70060 (7.867 sec)
INFO:tensorflow:lr = 0.00019781626 (7.868 sec)
INFO:tensorflow:global_step/sec: 12.6675
INFO:tensorflow:loss = 1.2597824, step = 70160 (7.890 sec)
INFO:tensorflow:lr = 0.00019761756 (7.892 sec)
INFO:tensorflow:global_step/sec: 13.1409
INFO:tensorflow:loss = 1.2613363, step = 70260 (7.612 sec)
INFO:tensorflow:lr = 0.00019741904 (7.611 sec)
INFO:tensorflow:global_step/sec: 12.9385
INFO:tensorflow:loss = 1.2608613, step = 70360 (7.732 sec)
INFO:tensorflow:lr = 0.00019722072 (7.731 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.20857
INFO:tensorflow:loss = 1.2632818, step = 70460 (12.177 sec)
INFO:tensorflow:lr = 0.0001970226 (12.180 sec)
INFO:tensorflow:global_step/sec: 12.7764
INFO:tensorflow:loss = 1.2565051, step = 70560 (7.830 sec)
INFO:tensorflow:lr = 0.0001968247 (7.829 sec)
INFO:tensorflow:global_step/sec: 12.8345
INFO:tensorflow:loss = 1.2527218, step = 70660 (7.791 sec)
INFO:tensorflow:lr = 0.00019662696 (7.791 sec)
INFO:tensorflow:global_step/sec: 12.5946
INFO:tensorflow:loss = 1.2486485, step = 70760 (7.944 sec)
INFO:tensorflow:lr = 0.00019642948 (7.941 sec)
INFO:tensorflow:global_step/sec: 12.8962
INFO:tensorflow:loss = 1.2574401, step = 70860 (7.755 sec)
INFO:tensorflow:lr = 0.00019623214 (7.755 sec)
INFO:tensorflow:global_step/sec: 12.8522
INFO:tensorflow:loss = 1.2536379, step = 70960 (7.778 sec)
INFO:tensorflow:lr = 0.000196035 (7.781 sec)
INFO:tensorflow:global_step/sec: 12.8556
INFO:tensorflow:loss = 1.253561, step = 71060 (7.776 sec)
INFO:tensorflow:lr = 0.0001958381 (7.774 sec)
INFO:tensorflow:global_step/sec: 12.6452
INFO:tensorflow:loss = 1.2638847, step = 71160 (7.909 sec)
INFO:tensorflow:lr = 0.00019564138 (7.908 sec)
INFO:tensorflow:global_step/sec: 12.881
INFO:tensorflow:loss = 1.2507195, step = 71260 (7.769 sec)
INFO:tensorflow:lr = 0.00019544485 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.9177
INFO:tensorflow:loss = 1.2516303, step = 71360 (7.740 sec)
INFO:tensorflow:lr = 0.00019524853 (7.740 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.34864
INFO:tensorflow:loss = 1.2554841, step = 71460 (11.978 sec)
INFO:tensorflow:lr = 0.00019505239 (11.978 sec)
INFO:tensorflow:global_step/sec: 12.7332
INFO:tensorflow:loss = 1.2557956, step = 71560 (7.848 sec)
INFO:tensorflow:lr = 0.00019485643 (7.855 sec)
INFO:tensorflow:global_step/sec: 12.8162
INFO:tensorflow:loss = 1.2516918, step = 71660 (7.802 sec)
INFO:tensorflow:lr = 0.00019466071 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.5072
INFO:tensorflow:loss = 1.2480005, step = 71760 (7.999 sec)
INFO:tensorflow:lr = 0.00019446517 (7.998 sec)
INFO:tensorflow:global_step/sec: 12.9678
INFO:tensorflow:loss = 1.2523248, step = 71860 (7.710 sec)
INFO:tensorflow:lr = 0.00019426983 (7.712 sec)
INFO:tensorflow:global_step/sec: 12.5132
INFO:tensorflow:loss = 1.2603774, step = 71960 (7.989 sec)
INFO:tensorflow:lr = 0.00019407466 (7.988 sec)
INFO:tensorflow:global_step/sec: 12.3631
INFO:tensorflow:loss = 1.2541709, step = 72060 (8.090 sec)
INFO:tensorflow:lr = 0.00019387972 (8.092 sec)
INFO:tensorflow:global_step/sec: 12.7245
INFO:tensorflow:loss = 1.2525417, step = 72160 (7.862 sec)
INFO:tensorflow:lr = 0.00019368496 (7.862 sec)
INFO:tensorflow:global_step/sec: 12.4976
INFO:tensorflow:loss = 1.2616658, step = 72260 (8.004 sec)
INFO:tensorflow:lr = 0.0001934904 (8.003 sec)
INFO:tensorflow:global_step/sec: 12.9216
INFO:tensorflow:loss = 1.2619153, step = 72360 (7.735 sec)
INFO:tensorflow:lr = 0.00019329603 (7.735 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.33511
INFO:tensorflow:loss = 1.2601154, step = 72460 (12.002 sec)
INFO:tensorflow:lr = 0.00019310186 (12.001 sec)
INFO:tensorflow:global_step/sec: 13.1451
INFO:tensorflow:loss = 1.251988, step = 72560 (7.604 sec)
INFO:tensorflow:lr = 0.00019290789 (7.604 sec)
INFO:tensorflow:global_step/sec: 12.6733
INFO:tensorflow:loss = 1.2546278, step = 72660 (7.888 sec)
INFO:tensorflow:lr = 0.0001927141 (7.888 sec)
INFO:tensorflow:global_step/sec: 12.3631
INFO:tensorflow:loss = 1.2519424, step = 72760 (8.088 sec)
INFO:tensorflow:lr = 0.00019252051 (8.088 sec)
INFO:tensorflow:global_step/sec: 12.6768
INFO:tensorflow:loss = 1.2527174, step = 72860 (7.889 sec)
INFO:tensorflow:lr = 0.00019232712 (7.889 sec)
INFO:tensorflow:global_step/sec: 12.743
INFO:tensorflow:loss = 1.2503885, step = 72960 (7.847 sec)
INFO:tensorflow:lr = 0.00019213393 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.433
INFO:tensorflow:loss = 1.2566782, step = 73060 (8.046 sec)
INFO:tensorflow:lr = 0.00019194091 (8.046 sec)
INFO:tensorflow:global_step/sec: 12.7119
INFO:tensorflow:loss = 1.2558047, step = 73160 (7.867 sec)
INFO:tensorflow:lr = 0.00019174813 (7.867 sec)
INFO:tensorflow:global_step/sec: 12.8101
INFO:tensorflow:loss = 1.2558652, step = 73260 (7.804 sec)
INFO:tensorflow:lr = 0.00019155549 (7.805 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.79918
INFO:tensorflow:loss = 1.2639738, step = 73360 (11.364 sec)
INFO:tensorflow:lr = 0.0001913631 (11.362 sec)
INFO:tensorflow:global_step/sec: 12.6976
INFO:tensorflow:loss = 1.2517492, step = 73460 (7.876 sec)
INFO:tensorflow:lr = 0.00019117084 (7.876 sec)
INFO:tensorflow:global_step/sec: 12.6726
INFO:tensorflow:loss = 1.2523484, step = 73560 (7.896 sec)
INFO:tensorflow:lr = 0.00019097881 (7.897 sec)
INFO:tensorflow:global_step/sec: 12.6952
INFO:tensorflow:loss = 1.250717, step = 73660 (7.871 sec)
INFO:tensorflow:lr = 0.00019078696 (7.871 sec)
INFO:tensorflow:global_step/sec: 12.9998
INFO:tensorflow:loss = 1.2526314, step = 73760 (7.693 sec)
INFO:tensorflow:lr = 0.00019059531 (7.694 sec)
INFO:tensorflow:global_step/sec: 12.9429
INFO:tensorflow:loss = 1.2553871, step = 73860 (7.726 sec)
INFO:tensorflow:lr = 0.00019040387 (7.728 sec)
INFO:tensorflow:global_step/sec: 12.7098
INFO:tensorflow:loss = 1.2495402, step = 73960 (7.873 sec)
INFO:tensorflow:lr = 0.00019021258 (7.869 sec)
INFO:tensorflow:global_step/sec: 12.7847
INFO:tensorflow:loss = 1.2583013, step = 74060 (7.818 sec)
INFO:tensorflow:lr = 0.00019002153 (7.820 sec)
INFO:tensorflow:global_step/sec: 12.606
INFO:tensorflow:loss = 1.2551463, step = 74160 (7.933 sec)
INFO:tensorflow:lr = 0.00018983062 (7.936 sec)
INFO:tensorflow:global_step/sec: 12.569
INFO:tensorflow:loss = 1.2499317, step = 74260 (7.961 sec)
INFO:tensorflow:lr = 0.00018963996 (7.956 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.31888
INFO:tensorflow:loss = 1.2556055, step = 74360 (10.732 sec)
INFO:tensorflow:lr = 0.00018944943 (10.731 sec)
INFO:tensorflow:global_step/sec: 12.5155
INFO:tensorflow:loss = 1.2517197, step = 74460 (7.990 sec)
INFO:tensorflow:lr = 0.00018925912 (7.991 sec)
INFO:tensorflow:global_step/sec: 12.8288
INFO:tensorflow:loss = 1.2531888, step = 74560 (7.789 sec)
INFO:tensorflow:lr = 0.00018906903 (7.788 sec)
INFO:tensorflow:global_step/sec: 12.5171
INFO:tensorflow:loss = 1.2516396, step = 74660 (7.989 sec)
INFO:tensorflow:lr = 0.0001888791 (7.993 sec)
INFO:tensorflow:global_step/sec: 12.7424
INFO:tensorflow:loss = 1.2480764, step = 74760 (7.848 sec)
INFO:tensorflow:lr = 0.00018868937 (7.844 sec)
INFO:tensorflow:global_step/sec: 12.5514
INFO:tensorflow:loss = 1.2534533, step = 74860 (7.970 sec)
INFO:tensorflow:lr = 0.00018849983 (7.970 sec)
INFO:tensorflow:global_step/sec: 12.8268
INFO:tensorflow:loss = 1.2513977, step = 74960 (7.800 sec)
INFO:tensorflow:lr = 0.00018831047 (7.799 sec)
INFO:tensorflow:global_step/sec: 12.8144
INFO:tensorflow:loss = 1.2506344, step = 75060 (7.798 sec)
INFO:tensorflow:lr = 0.0001881213 (7.802 sec)
INFO:tensorflow:global_step/sec: 12.7008
INFO:tensorflow:loss = 1.2517682, step = 75160 (7.874 sec)
INFO:tensorflow:lr = 0.00018793234 (7.872 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2503
INFO:tensorflow:loss = 1.2574689, step = 75260 (8.171 sec)
INFO:tensorflow:lr = 0.00018774354 (8.176 sec)
INFO:tensorflow:global_step/sec: 10.6335
INFO:tensorflow:loss = 1.26586, step = 75360 (9.400 sec)
INFO:tensorflow:lr = 0.00018755496 (9.393 sec)
INFO:tensorflow:global_step/sec: 12.973
INFO:tensorflow:loss = 1.2528296, step = 75460 (7.707 sec)
INFO:tensorflow:lr = 0.00018736653 (7.709 sec)
INFO:tensorflow:global_step/sec: 12.2656
INFO:tensorflow:loss = 1.2583458, step = 75560 (8.154 sec)
INFO:tensorflow:lr = 0.00018717833 (8.153 sec)
INFO:tensorflow:global_step/sec: 12.8809
INFO:tensorflow:loss = 1.2474772, step = 75660 (7.759 sec)
INFO:tensorflow:lr = 0.00018699029 (7.763 sec)
INFO:tensorflow:global_step/sec: 12.8353
INFO:tensorflow:loss = 1.2567656, step = 75760 (7.796 sec)
INFO:tensorflow:lr = 0.00018680248 (7.791 sec)
INFO:tensorflow:global_step/sec: 12.8607
INFO:tensorflow:loss = 1.2529895, step = 75860 (7.776 sec)
INFO:tensorflow:lr = 0.00018661484 (7.777 sec)
INFO:tensorflow:global_step/sec: 12.4908
INFO:tensorflow:loss = 1.255029, step = 75960 (8.005 sec)
INFO:tensorflow:lr = 0.00018642735 (8.006 sec)
INFO:tensorflow:global_step/sec: 12.4031
INFO:tensorflow:loss = 1.2508453, step = 76060 (8.062 sec)
INFO:tensorflow:lr = 0.00018624008 (8.060 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2921
INFO:tensorflow:loss = 1.2630798, step = 76160 (8.134 sec)
INFO:tensorflow:lr = 0.000186053 (8.136 sec)
INFO:tensorflow:global_step/sec: 10.9934
INFO:tensorflow:loss = 1.255268, step = 76260 (9.093 sec)
INFO:tensorflow:lr = 0.00018586611 (9.093 sec)
INFO:tensorflow:global_step/sec: 12.8284
INFO:tensorflow:loss = 1.2487235, step = 76360 (7.795 sec)
INFO:tensorflow:lr = 0.00018567941 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.6785
INFO:tensorflow:loss = 1.2549771, step = 76460 (7.893 sec)
INFO:tensorflow:lr = 0.0001854929 (7.893 sec)
INFO:tensorflow:global_step/sec: 12.5547
INFO:tensorflow:loss = 1.256425, step = 76560 (7.963 sec)
INFO:tensorflow:lr = 0.00018530655 (7.964 sec)
INFO:tensorflow:global_step/sec: 12.5549
INFO:tensorflow:loss = 1.2490952, step = 76660 (7.962 sec)
INFO:tensorflow:lr = 0.00018512041 (7.961 sec)
INFO:tensorflow:global_step/sec: 12.831
INFO:tensorflow:loss = 1.2511157, step = 76760 (7.792 sec)
INFO:tensorflow:lr = 0.00018493446 (7.792 sec)
INFO:tensorflow:global_step/sec: 12.6078
INFO:tensorflow:loss = 1.2497431, step = 76860 (7.937 sec)
INFO:tensorflow:lr = 0.00018474869 (7.938 sec)
INFO:tensorflow:global_step/sec: 12.6514
INFO:tensorflow:loss = 1.2566417, step = 76960 (7.902 sec)
INFO:tensorflow:lr = 0.00018456309 (7.901 sec)
INFO:tensorflow:global_step/sec: 12.9655
INFO:tensorflow:loss = 1.2616142, step = 77060 (7.713 sec)
INFO:tensorflow:lr = 0.00018437768 (7.715 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.9968
INFO:tensorflow:loss = 1.2498459, step = 77160 (8.342 sec)
INFO:tensorflow:lr = 0.00018419247 (8.347 sec)
INFO:tensorflow:global_step/sec: 11.3735
INFO:tensorflow:loss = 1.2483019, step = 77260 (8.789 sec)
INFO:tensorflow:lr = 0.00018400745 (8.781 sec)
INFO:tensorflow:global_step/sec: 12.7229
INFO:tensorflow:loss = 1.2514182, step = 77360 (7.857 sec)
INFO:tensorflow:lr = 0.00018382262 (7.858 sec)
INFO:tensorflow:global_step/sec: 12.6732
INFO:tensorflow:loss = 1.2530181, step = 77460 (7.893 sec)
INFO:tensorflow:lr = 0.00018363795 (7.894 sec)
INFO:tensorflow:global_step/sec: 12.605
INFO:tensorflow:loss = 1.2535021, step = 77560 (7.931 sec)
INFO:tensorflow:lr = 0.00018345349 (7.930 sec)
INFO:tensorflow:global_step/sec: 12.9194
INFO:tensorflow:loss = 1.2538028, step = 77660 (7.743 sec)
INFO:tensorflow:lr = 0.00018326919 (7.742 sec)
INFO:tensorflow:global_step/sec: 12.62
INFO:tensorflow:loss = 1.2521385, step = 77760 (7.923 sec)
INFO:tensorflow:lr = 0.00018308511 (7.923 sec)
INFO:tensorflow:global_step/sec: 12.8095
INFO:tensorflow:loss = 1.252689, step = 77860 (7.806 sec)
INFO:tensorflow:lr = 0.0001829012 (7.807 sec)
INFO:tensorflow:global_step/sec: 12.977
INFO:tensorflow:loss = 1.2580132, step = 77960 (7.707 sec)
INFO:tensorflow:lr = 0.00018271746 (7.707 sec)
INFO:tensorflow:global_step/sec: 12.992
INFO:tensorflow:loss = 1.2574784, step = 78060 (7.695 sec)
INFO:tensorflow:lr = 0.00018253391 (7.695 sec)
INFO:tensorflow:global_step/sec: 12.6106
INFO:tensorflow:loss = 1.2529795, step = 78160 (7.931 sec)
INFO:tensorflow:lr = 0.00018235056 (7.930 sec)
INFO:tensorflow:Saving checkpoints for 78240 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2499174.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-78240
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-78240
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.717
INFO:tensorflow:Best (Exact Match) Accuracy: 0.717
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-78240
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 78240 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2590173, step = 78240
INFO:tensorflow:lr = 0.000182204
INFO:tensorflow:global_step/sec: 10.2335
INFO:tensorflow:loss = 1.2496594, step = 78340 (9.780 sec)
INFO:tensorflow:lr = 0.00018202097 (9.780 sec)
INFO:tensorflow:global_step/sec: 12.6641
INFO:tensorflow:loss = 1.2591709, step = 78440 (7.896 sec)
INFO:tensorflow:lr = 0.00018183814 (7.896 sec)
INFO:tensorflow:global_step/sec: 12.8297
INFO:tensorflow:loss = 1.2547472, step = 78540 (7.788 sec)
INFO:tensorflow:lr = 0.00018165546 (7.792 sec)
INFO:tensorflow:global_step/sec: 12.8503
INFO:tensorflow:loss = 1.2477771, step = 78640 (7.788 sec)
INFO:tensorflow:lr = 0.00018147299 (7.785 sec)
INFO:tensorflow:global_step/sec: 12.4826
INFO:tensorflow:loss = 1.250107, step = 78740 (8.005 sec)
INFO:tensorflow:lr = 0.0001812907 (8.005 sec)
INFO:tensorflow:global_step/sec: 12.6461
INFO:tensorflow:loss = 1.2553511, step = 78840 (7.909 sec)
INFO:tensorflow:lr = 0.00018110858 (7.909 sec)
INFO:tensorflow:global_step/sec: 12.8929
INFO:tensorflow:loss = 1.2516081, step = 78940 (7.762 sec)
INFO:tensorflow:lr = 0.00018092665 (7.763 sec)
INFO:tensorflow:global_step/sec: 12.6101
INFO:tensorflow:loss = 1.2541536, step = 79040 (7.927 sec)
INFO:tensorflow:lr = 0.00018074492 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.4696
INFO:tensorflow:loss = 1.2553225, step = 79140 (8.015 sec)
INFO:tensorflow:lr = 0.00018056334 (8.015 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.35453
INFO:tensorflow:loss = 1.2502196, step = 79240 (11.971 sec)
INFO:tensorflow:lr = 0.00018038196 (11.971 sec)
INFO:tensorflow:global_step/sec: 12.639
INFO:tensorflow:loss = 1.2486528, step = 79340 (7.912 sec)
INFO:tensorflow:lr = 0.00018020078 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.694
INFO:tensorflow:loss = 1.248358, step = 79440 (7.883 sec)
INFO:tensorflow:lr = 0.00018001975 (7.883 sec)
INFO:tensorflow:global_step/sec: 12.6918
INFO:tensorflow:loss = 1.2552594, step = 79540 (7.873 sec)
INFO:tensorflow:lr = 0.00017983893 (7.873 sec)
INFO:tensorflow:global_step/sec: 12.7387
INFO:tensorflow:loss = 1.2540479, step = 79640 (7.857 sec)
INFO:tensorflow:lr = 0.00017965824 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.7917
INFO:tensorflow:loss = 1.2457458, step = 79740 (7.811 sec)
INFO:tensorflow:lr = 0.00017947779 (7.805 sec)
INFO:tensorflow:global_step/sec: 12.4802
INFO:tensorflow:loss = 1.2563258, step = 79840 (8.015 sec)
INFO:tensorflow:lr = 0.0001792975 (8.016 sec)
INFO:tensorflow:global_step/sec: 12.8425
INFO:tensorflow:loss = 1.2508807, step = 79940 (7.790 sec)
INFO:tensorflow:lr = 0.0001791174 (7.789 sec)
INFO:tensorflow:global_step/sec: 12.7452
INFO:tensorflow:loss = 1.2514696, step = 80040 (7.845 sec)
INFO:tensorflow:lr = 0.00017893747 (7.844 sec)
INFO:tensorflow:global_step/sec: 12.8878
INFO:tensorflow:loss = 1.2509739, step = 80140 (7.755 sec)
INFO:tensorflow:lr = 0.0001787577 (7.757 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.38154
INFO:tensorflow:loss = 1.2584447, step = 80240 (11.935 sec)
INFO:tensorflow:lr = 0.00017857815 (11.934 sec)
INFO:tensorflow:global_step/sec: 12.6567
INFO:tensorflow:loss = 1.2512501, step = 80340 (7.896 sec)
INFO:tensorflow:lr = 0.00017839875 (7.901 sec)
INFO:tensorflow:global_step/sec: 12.7351
INFO:tensorflow:loss = 1.2498204, step = 80440 (7.853 sec)
INFO:tensorflow:lr = 0.00017821955 (7.850 sec)
INFO:tensorflow:global_step/sec: 12.8998
INFO:tensorflow:loss = 1.254194, step = 80540 (7.757 sec)
INFO:tensorflow:lr = 0.00017804053 (7.756 sec)
INFO:tensorflow:global_step/sec: 12.6751
INFO:tensorflow:loss = 1.2590884, step = 80640 (7.887 sec)
INFO:tensorflow:lr = 0.00017786169 (7.889 sec)
INFO:tensorflow:global_step/sec: 12.8078
INFO:tensorflow:loss = 1.2537712, step = 80740 (7.811 sec)
INFO:tensorflow:lr = 0.000177683 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.8097
INFO:tensorflow:loss = 1.2545077, step = 80840 (7.800 sec)
INFO:tensorflow:lr = 0.00017750452 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.8288
INFO:tensorflow:loss = 1.24793, step = 80940 (7.794 sec)
INFO:tensorflow:lr = 0.00017732622 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.9341
INFO:tensorflow:loss = 1.2493593, step = 81040 (7.732 sec)
INFO:tensorflow:lr = 0.00017714809 (7.731 sec)
INFO:tensorflow:global_step/sec: 12.6921
INFO:tensorflow:loss = 1.2523218, step = 81140 (7.879 sec)
INFO:tensorflow:lr = 0.00017697013 (7.880 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.29548
INFO:tensorflow:loss = 1.2478547, step = 81240 (12.054 sec)
INFO:tensorflow:lr = 0.00017679235 (12.054 sec)
INFO:tensorflow:global_step/sec: 12.8312
INFO:tensorflow:loss = 1.2478245, step = 81340 (7.797 sec)
INFO:tensorflow:lr = 0.00017661476 (7.798 sec)
INFO:tensorflow:global_step/sec: 12.5046
INFO:tensorflow:loss = 1.2513237, step = 81440 (7.999 sec)
INFO:tensorflow:lr = 0.00017643734 (7.999 sec)
INFO:tensorflow:global_step/sec: 13.0428
INFO:tensorflow:loss = 1.2500213, step = 81540 (7.664 sec)
INFO:tensorflow:lr = 0.00017626012 (7.665 sec)
INFO:tensorflow:global_step/sec: 12.8603
INFO:tensorflow:loss = 1.2503232, step = 81640 (7.774 sec)
INFO:tensorflow:lr = 0.00017608305 (7.772 sec)
INFO:tensorflow:global_step/sec: 12.8574
INFO:tensorflow:loss = 1.2482414, step = 81740 (7.781 sec)
INFO:tensorflow:lr = 0.00017590619 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.2935
INFO:tensorflow:loss = 1.2473918, step = 81840 (8.131 sec)
INFO:tensorflow:lr = 0.00017572947 (8.131 sec)
INFO:tensorflow:global_step/sec: 12.9845
INFO:tensorflow:loss = 1.2510718, step = 81940 (7.701 sec)
INFO:tensorflow:lr = 0.00017555297 (7.702 sec)
INFO:tensorflow:global_step/sec: 12.6733
INFO:tensorflow:loss = 1.2469774, step = 82040 (7.891 sec)
INFO:tensorflow:lr = 0.00017537661 (7.892 sec)
INFO:tensorflow:global_step/sec: 12.6176
INFO:tensorflow:loss = 1.2501423, step = 82140 (7.925 sec)
INFO:tensorflow:lr = 0.00017520043 (7.928 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.49764
INFO:tensorflow:loss = 1.2507443, step = 82240 (11.768 sec)
INFO:tensorflow:lr = 0.00017502444 (11.764 sec)
INFO:tensorflow:global_step/sec: 12.5804
INFO:tensorflow:loss = 1.2484735, step = 82340 (7.952 sec)
INFO:tensorflow:lr = 0.00017484862 (7.953 sec)
INFO:tensorflow:global_step/sec: 13.0737
INFO:tensorflow:loss = 1.250423, step = 82440 (7.648 sec)
INFO:tensorflow:lr = 0.000174673 (7.648 sec)
INFO:tensorflow:global_step/sec: 12.6908
INFO:tensorflow:loss = 1.2670717, step = 82540 (7.882 sec)
INFO:tensorflow:lr = 0.00017449753 (7.882 sec)
INFO:tensorflow:global_step/sec: 12.838
INFO:tensorflow:loss = 1.249461, step = 82640 (7.791 sec)
INFO:tensorflow:lr = 0.00017432224 (7.793 sec)
INFO:tensorflow:global_step/sec: 12.7286
INFO:tensorflow:loss = 1.2500602, step = 82740 (7.849 sec)
INFO:tensorflow:lr = 0.0001741471 (7.847 sec)
INFO:tensorflow:global_step/sec: 12.8635
INFO:tensorflow:loss = 1.250238, step = 82840 (7.776 sec)
INFO:tensorflow:lr = 0.00017397219 (7.778 sec)
INFO:tensorflow:global_step/sec: 12.5899
INFO:tensorflow:loss = 1.2476375, step = 82940 (7.945 sec)
INFO:tensorflow:lr = 0.00017379744 (7.943 sec)
INFO:tensorflow:global_step/sec: 12.8872
INFO:tensorflow:loss = 1.2561243, step = 83040 (7.760 sec)
INFO:tensorflow:lr = 0.00017362286 (7.760 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.60542
INFO:tensorflow:loss = 1.249593, step = 83140 (11.621 sec)
INFO:tensorflow:lr = 0.00017344844 (11.621 sec)
INFO:tensorflow:global_step/sec: 13.0203
INFO:tensorflow:loss = 1.2486179, step = 83240 (7.676 sec)
INFO:tensorflow:lr = 0.00017327421 (7.676 sec)
INFO:tensorflow:global_step/sec: 12.5732
INFO:tensorflow:loss = 1.2553233, step = 83340 (7.958 sec)
INFO:tensorflow:lr = 0.00017310014 (7.959 sec)
INFO:tensorflow:global_step/sec: 12.7835
INFO:tensorflow:loss = 1.2564698, step = 83440 (7.822 sec)
INFO:tensorflow:lr = 0.00017292627 (7.821 sec)
INFO:tensorflow:global_step/sec: 12.9991
INFO:tensorflow:loss = 1.2600302, step = 83540 (7.694 sec)
INFO:tensorflow:lr = 0.00017275255 (7.694 sec)
INFO:tensorflow:global_step/sec: 12.6979
INFO:tensorflow:loss = 1.2510674, step = 83640 (7.870 sec)
INFO:tensorflow:lr = 0.000172579 (7.873 sec)
INFO:tensorflow:global_step/sec: 12.7094
INFO:tensorflow:loss = 1.2540224, step = 83740 (7.869 sec)
INFO:tensorflow:lr = 0.00017240565 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.7121
INFO:tensorflow:loss = 1.2510121, step = 83840 (7.869 sec)
INFO:tensorflow:lr = 0.00017223245 (7.869 sec)
INFO:tensorflow:global_step/sec: 12.6853
INFO:tensorflow:loss = 1.2614245, step = 83940 (7.885 sec)
INFO:tensorflow:lr = 0.00017205944 (7.884 sec)
INFO:tensorflow:global_step/sec: 12.7461
INFO:tensorflow:loss = 1.2500675, step = 84040 (7.844 sec)
INFO:tensorflow:lr = 0.00017188663 (7.847 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.15836
INFO:tensorflow:loss = 1.259658, step = 84140 (10.922 sec)
INFO:tensorflow:lr = 0.00017171395 (10.920 sec)
INFO:tensorflow:global_step/sec: 13.0493
INFO:tensorflow:loss = 1.2597753, step = 84240 (7.659 sec)
INFO:tensorflow:lr = 0.00017154145 (7.658 sec)
INFO:tensorflow:global_step/sec: 12.8351
INFO:tensorflow:loss = 1.2478659, step = 84340 (7.792 sec)
INFO:tensorflow:lr = 0.00017136915 (7.792 sec)
INFO:tensorflow:global_step/sec: 12.6189
INFO:tensorflow:loss = 1.2543241, step = 84440 (7.923 sec)
INFO:tensorflow:lr = 0.00017119701 (7.922 sec)
INFO:tensorflow:global_step/sec: 12.8493
INFO:tensorflow:loss = 1.2495148, step = 84540 (7.783 sec)
INFO:tensorflow:lr = 0.00017102502 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.5098
INFO:tensorflow:loss = 1.2590244, step = 84640 (7.998 sec)
INFO:tensorflow:lr = 0.00017085322 (7.995 sec)
INFO:tensorflow:global_step/sec: 12.8627
INFO:tensorflow:loss = 1.2492456, step = 84740 (7.770 sec)
INFO:tensorflow:lr = 0.0001706816 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.8319
INFO:tensorflow:loss = 1.2496346, step = 84840 (7.796 sec)
INFO:tensorflow:lr = 0.00017051013 (7.797 sec)
INFO:tensorflow:global_step/sec: 12.6574
INFO:tensorflow:loss = 1.2487538, step = 84940 (7.898 sec)
INFO:tensorflow:lr = 0.00017033887 (7.901 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.3381
INFO:tensorflow:loss = 1.2556379, step = 85040 (8.113 sec)
INFO:tensorflow:lr = 0.00017016775 (8.114 sec)
INFO:tensorflow:global_step/sec: 10.7431
INFO:tensorflow:loss = 1.2477072, step = 85140 (9.305 sec)
INFO:tensorflow:lr = 0.00016999681 (9.300 sec)
INFO:tensorflow:global_step/sec: 13.1212
INFO:tensorflow:loss = 1.2478539, step = 85240 (7.621 sec)
INFO:tensorflow:lr = 0.00016982605 (7.625 sec)
INFO:tensorflow:global_step/sec: 13.0129
INFO:tensorflow:loss = 1.2585075, step = 85340 (7.680 sec)
INFO:tensorflow:lr = 0.00016965547 (7.676 sec)
INFO:tensorflow:global_step/sec: 12.9017
INFO:tensorflow:loss = 1.2566411, step = 85440 (7.750 sec)
INFO:tensorflow:lr = 0.00016948502 (7.750 sec)
INFO:tensorflow:global_step/sec: 12.7359
INFO:tensorflow:loss = 1.2534791, step = 85540 (7.857 sec)
INFO:tensorflow:lr = 0.00016931478 (7.858 sec)
INFO:tensorflow:global_step/sec: 12.2424
INFO:tensorflow:loss = 1.2601684, step = 85640 (8.168 sec)
INFO:tensorflow:lr = 0.00016914468 (8.168 sec)
INFO:tensorflow:global_step/sec: 12.5107
INFO:tensorflow:loss = 1.2539284, step = 85740 (7.995 sec)
INFO:tensorflow:lr = 0.00016897477 (7.995 sec)
INFO:tensorflow:global_step/sec: 12.9542
INFO:tensorflow:loss = 1.2505609, step = 85840 (7.713 sec)
INFO:tensorflow:lr = 0.00016880504 (7.712 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.6341
INFO:tensorflow:loss = 1.2487622, step = 85940 (7.922 sec)
INFO:tensorflow:lr = 0.00016863547 (7.923 sec)
INFO:tensorflow:global_step/sec: 10.8304
INFO:tensorflow:loss = 1.2539859, step = 86040 (9.231 sec)
INFO:tensorflow:lr = 0.00016846608 (9.230 sec)
INFO:tensorflow:global_step/sec: 12.7512
INFO:tensorflow:loss = 1.2534391, step = 86140 (7.844 sec)
INFO:tensorflow:lr = 0.00016829684 (7.846 sec)
INFO:tensorflow:global_step/sec: 12.7686
INFO:tensorflow:loss = 1.2543765, step = 86240 (7.829 sec)
INFO:tensorflow:lr = 0.00016812778 (7.827 sec)
INFO:tensorflow:global_step/sec: 12.956
INFO:tensorflow:loss = 1.2502369, step = 86340 (7.719 sec)
INFO:tensorflow:lr = 0.00016795889 (7.719 sec)
INFO:tensorflow:global_step/sec: 12.8613
INFO:tensorflow:loss = 1.2478958, step = 86440 (7.771 sec)
INFO:tensorflow:lr = 0.00016779019 (7.772 sec)
INFO:tensorflow:global_step/sec: 13.1076
INFO:tensorflow:loss = 1.2502476, step = 86540 (7.628 sec)
INFO:tensorflow:lr = 0.00016762162 (7.635 sec)
INFO:tensorflow:global_step/sec: 12.6676
INFO:tensorflow:loss = 1.2501, step = 86640 (7.895 sec)
INFO:tensorflow:lr = 0.00016745325 (7.888 sec)
INFO:tensorflow:global_step/sec: 12.7157
INFO:tensorflow:loss = 1.2478064, step = 86740 (7.863 sec)
INFO:tensorflow:lr = 0.00016728503 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.7067
INFO:tensorflow:loss = 1.2639413, step = 86840 (7.870 sec)
INFO:tensorflow:lr = 0.00016711699 (7.874 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0525
INFO:tensorflow:loss = 1.252271, step = 86940 (8.303 sec)
INFO:tensorflow:lr = 0.00016694913 (8.304 sec)
INFO:tensorflow:global_step/sec: 11.429
INFO:tensorflow:loss = 1.252577, step = 87040 (8.749 sec)
INFO:tensorflow:lr = 0.0001667814 (8.743 sec)
INFO:tensorflow:global_step/sec: 12.8618
INFO:tensorflow:loss = 1.2451332, step = 87140 (7.772 sec)
INFO:tensorflow:lr = 0.00016661389 (7.774 sec)
INFO:tensorflow:global_step/sec: 13.0815
INFO:tensorflow:loss = 1.250391, step = 87240 (7.649 sec)
INFO:tensorflow:lr = 0.00016644651 (7.649 sec)
INFO:tensorflow:global_step/sec: 12.8
INFO:tensorflow:loss = 1.250594, step = 87340 (7.811 sec)
INFO:tensorflow:lr = 0.00016627931 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.9375
INFO:tensorflow:loss = 1.2503841, step = 87440 (7.726 sec)
INFO:tensorflow:lr = 0.00016611228 (7.727 sec)
INFO:tensorflow:global_step/sec: 12.8328
INFO:tensorflow:loss = 1.2486295, step = 87540 (7.795 sec)
INFO:tensorflow:lr = 0.00016594543 (7.794 sec)
INFO:tensorflow:global_step/sec: 12.5163
INFO:tensorflow:loss = 1.2504075, step = 87640 (7.985 sec)
INFO:tensorflow:lr = 0.00016577872 (7.989 sec)
INFO:tensorflow:global_step/sec: 13.0312
INFO:tensorflow:loss = 1.2518604, step = 87740 (7.674 sec)
INFO:tensorflow:lr = 0.00016561219 (7.671 sec)
INFO:tensorflow:global_step/sec: 12.6941
INFO:tensorflow:loss = 1.2477338, step = 87840 (7.878 sec)
INFO:tensorflow:lr = 0.00016544582 (7.878 sec)
INFO:tensorflow:global_step/sec: 12.3306
INFO:tensorflow:loss = 1.2489831, step = 87940 (8.110 sec)
INFO:tensorflow:lr = 0.00016527963 (8.109 sec)
INFO:tensorflow:Saving checkpoints for 88020 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2430813.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-88020
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-88020
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.721
INFO:tensorflow:Best (Exact Match) Accuracy: 0.721
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-88020
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 88020 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2478541, step = 88020
INFO:tensorflow:lr = 0.00016514679
INFO:tensorflow:global_step/sec: 10.07
INFO:tensorflow:loss = 1.2509733, step = 88120 (9.934 sec)
INFO:tensorflow:lr = 0.0001649809 (9.935 sec)
INFO:tensorflow:global_step/sec: 12.5025
INFO:tensorflow:loss = 1.249775, step = 88220 (7.999 sec)
INFO:tensorflow:lr = 0.00016481518 (7.996 sec)
INFO:tensorflow:global_step/sec: 12.7872
INFO:tensorflow:loss = 1.2507293, step = 88320 (7.825 sec)
INFO:tensorflow:lr = 0.00016464961 (7.831 sec)
INFO:tensorflow:global_step/sec: 12.7513
INFO:tensorflow:loss = 1.2496305, step = 88420 (7.836 sec)
INFO:tensorflow:lr = 0.00016448423 (7.834 sec)
INFO:tensorflow:global_step/sec: 12.9031
INFO:tensorflow:loss = 1.2485484, step = 88520 (7.751 sec)
INFO:tensorflow:lr = 0.00016431899 (7.748 sec)
INFO:tensorflow:global_step/sec: 12.803
INFO:tensorflow:loss = 1.2538512, step = 88620 (7.814 sec)
INFO:tensorflow:lr = 0.00016415393 (7.814 sec)
INFO:tensorflow:global_step/sec: 12.4804
INFO:tensorflow:loss = 1.2560003, step = 88720 (8.008 sec)
INFO:tensorflow:lr = 0.00016398903 (8.010 sec)
INFO:tensorflow:global_step/sec: 12.8263
INFO:tensorflow:loss = 1.2487856, step = 88820 (7.796 sec)
INFO:tensorflow:lr = 0.0001638243 (7.794 sec)
INFO:tensorflow:global_step/sec: 12.9732
INFO:tensorflow:loss = 1.2564486, step = 88920 (7.708 sec)
INFO:tensorflow:lr = 0.00016365973 (7.708 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.3463
INFO:tensorflow:loss = 1.248512, step = 89020 (11.987 sec)
INFO:tensorflow:lr = 0.00016349532 (11.987 sec)
INFO:tensorflow:global_step/sec: 12.8714
INFO:tensorflow:loss = 1.2499951, step = 89120 (7.770 sec)
INFO:tensorflow:lr = 0.0001633311 (7.771 sec)
INFO:tensorflow:global_step/sec: 12.754
INFO:tensorflow:loss = 1.2510864, step = 89220 (7.839 sec)
INFO:tensorflow:lr = 0.00016316702 (7.839 sec)
INFO:tensorflow:global_step/sec: 12.7582
INFO:tensorflow:loss = 1.2505963, step = 89320 (7.835 sec)
INFO:tensorflow:lr = 0.00016300312 (7.836 sec)
INFO:tensorflow:global_step/sec: 12.6331
INFO:tensorflow:loss = 1.2495183, step = 89420 (7.915 sec)
INFO:tensorflow:lr = 0.00016283937 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.6353
INFO:tensorflow:loss = 1.250917, step = 89520 (7.914 sec)
INFO:tensorflow:lr = 0.0001626758 (7.917 sec)
INFO:tensorflow:global_step/sec: 12.993
INFO:tensorflow:loss = 1.2484233, step = 89620 (7.699 sec)
INFO:tensorflow:lr = 0.0001625124 (7.696 sec)
INFO:tensorflow:global_step/sec: 12.8745
INFO:tensorflow:loss = 1.2567275, step = 89720 (7.770 sec)
INFO:tensorflow:lr = 0.00016234916 (7.770 sec)
INFO:tensorflow:global_step/sec: 12.7192
INFO:tensorflow:loss = 1.2655329, step = 89820 (7.862 sec)
INFO:tensorflow:lr = 0.00016218606 (7.862 sec)
INFO:tensorflow:global_step/sec: 12.7579
INFO:tensorflow:loss = 1.2497483, step = 89920 (7.833 sec)
INFO:tensorflow:lr = 0.00016202313 (7.834 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.35646
INFO:tensorflow:loss = 1.2529335, step = 90020 (11.966 sec)
INFO:tensorflow:lr = 0.00016186039 (11.965 sec)
INFO:tensorflow:global_step/sec: 12.8463
INFO:tensorflow:loss = 1.248037, step = 90120 (7.789 sec)
INFO:tensorflow:lr = 0.00016169778 (7.790 sec)
INFO:tensorflow:global_step/sec: 12.7832
INFO:tensorflow:loss = 1.2473701, step = 90220 (7.818 sec)
INFO:tensorflow:lr = 0.00016153537 (7.819 sec)
INFO:tensorflow:global_step/sec: 12.9166
INFO:tensorflow:loss = 1.2491748, step = 90320 (7.747 sec)
INFO:tensorflow:lr = 0.00016137307 (7.747 sec)
INFO:tensorflow:global_step/sec: 12.7024
INFO:tensorflow:loss = 1.2531077, step = 90420 (7.871 sec)
INFO:tensorflow:lr = 0.00016121099 (7.871 sec)
INFO:tensorflow:global_step/sec: 12.6706
INFO:tensorflow:loss = 1.2527239, step = 90520 (7.894 sec)
INFO:tensorflow:lr = 0.00016104903 (7.893 sec)
INFO:tensorflow:global_step/sec: 12.5734
INFO:tensorflow:loss = 1.2484795, step = 90620 (7.952 sec)
INFO:tensorflow:lr = 0.00016088727 (7.953 sec)
INFO:tensorflow:global_step/sec: 12.621
INFO:tensorflow:loss = 1.2495238, step = 90720 (7.924 sec)
INFO:tensorflow:lr = 0.00016072566 (7.924 sec)
INFO:tensorflow:global_step/sec: 12.8078
INFO:tensorflow:loss = 1.2501513, step = 90820 (7.802 sec)
INFO:tensorflow:lr = 0.00016056419 (7.803 sec)
INFO:tensorflow:global_step/sec: 12.96
INFO:tensorflow:loss = 1.2493507, step = 90920 (7.716 sec)
INFO:tensorflow:lr = 0.0001604029 (7.715 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.41387
INFO:tensorflow:loss = 1.2496051, step = 91020 (11.889 sec)
INFO:tensorflow:lr = 0.00016024178 (11.897 sec)
INFO:tensorflow:global_step/sec: 12.9541
INFO:tensorflow:loss = 1.247105, step = 91120 (7.718 sec)
INFO:tensorflow:lr = 0.00016008082 (7.710 sec)
INFO:tensorflow:global_step/sec: 12.7781
INFO:tensorflow:loss = 1.2484589, step = 91220 (7.825 sec)
INFO:tensorflow:lr = 0.00015992 (7.826 sec)
INFO:tensorflow:global_step/sec: 13.0741
INFO:tensorflow:loss = 1.2535778, step = 91320 (7.650 sec)
INFO:tensorflow:lr = 0.00015975937 (7.649 sec)
INFO:tensorflow:global_step/sec: 12.5744
INFO:tensorflow:loss = 1.255234, step = 91420 (7.952 sec)
INFO:tensorflow:lr = 0.00015959887 (7.953 sec)
INFO:tensorflow:global_step/sec: 12.4252
INFO:tensorflow:loss = 1.2474556, step = 91520 (8.050 sec)
INFO:tensorflow:lr = 0.00015943855 (8.050 sec)
INFO:tensorflow:global_step/sec: 12.5663
INFO:tensorflow:loss = 1.245488, step = 91620 (7.954 sec)
INFO:tensorflow:lr = 0.0001592784 (7.955 sec)
INFO:tensorflow:global_step/sec: 12.7127
INFO:tensorflow:loss = 1.2660935, step = 91720 (7.866 sec)
INFO:tensorflow:lr = 0.0001591184 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.5399
INFO:tensorflow:loss = 1.2542043, step = 91820 (7.975 sec)
INFO:tensorflow:lr = 0.00015895854 (7.976 sec)
INFO:tensorflow:global_step/sec: 12.8243
INFO:tensorflow:loss = 1.250056, step = 91920 (7.798 sec)
INFO:tensorflow:lr = 0.00015879888 (7.806 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.35817
INFO:tensorflow:loss = 1.2492906, step = 92020 (11.971 sec)
INFO:tensorflow:lr = 0.00015863935 (11.962 sec)
INFO:tensorflow:global_step/sec: 12.9871
INFO:tensorflow:loss = 1.257855, step = 92120 (7.695 sec)
INFO:tensorflow:lr = 0.00015848 (7.695 sec)
INFO:tensorflow:global_step/sec: 12.8269
INFO:tensorflow:loss = 1.2487328, step = 92220 (7.795 sec)
INFO:tensorflow:lr = 0.0001583208 (7.799 sec)
INFO:tensorflow:global_step/sec: 12.7098
INFO:tensorflow:loss = 1.2490268, step = 92320 (7.870 sec)
INFO:tensorflow:lr = 0.00015816175 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.9235
INFO:tensorflow:loss = 1.2516989, step = 92420 (7.740 sec)
INFO:tensorflow:lr = 0.00015800289 (7.739 sec)
INFO:tensorflow:global_step/sec: 13.0381
INFO:tensorflow:loss = 1.2501904, step = 92520 (7.668 sec)
INFO:tensorflow:lr = 0.00015784417 (7.668 sec)
INFO:tensorflow:global_step/sec: 12.812
INFO:tensorflow:loss = 1.2477765, step = 92620 (7.809 sec)
INFO:tensorflow:lr = 0.00015768562 (7.810 sec)
INFO:tensorflow:global_step/sec: 12.697
INFO:tensorflow:loss = 1.2476797, step = 92720 (7.873 sec)
INFO:tensorflow:lr = 0.0001575272 (7.875 sec)
INFO:tensorflow:global_step/sec: 12.7559
INFO:tensorflow:loss = 1.2522918, step = 92820 (7.838 sec)
INFO:tensorflow:lr = 0.00015736897 (7.837 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.66571
INFO:tensorflow:loss = 1.2524945, step = 92920 (11.542 sec)
INFO:tensorflow:lr = 0.00015721089 (11.544 sec)
INFO:tensorflow:global_step/sec: 12.7391
INFO:tensorflow:loss = 1.2476618, step = 93020 (7.850 sec)
INFO:tensorflow:lr = 0.00015705297 (7.846 sec)
INFO:tensorflow:global_step/sec: 12.8889
INFO:tensorflow:loss = 1.2466986, step = 93120 (7.763 sec)
INFO:tensorflow:lr = 0.0001568952 (7.763 sec)
INFO:tensorflow:global_step/sec: 12.5188
INFO:tensorflow:loss = 1.2498853, step = 93220 (7.981 sec)
INFO:tensorflow:lr = 0.0001567376 (7.981 sec)
INFO:tensorflow:global_step/sec: 12.814
INFO:tensorflow:loss = 1.2529539, step = 93320 (7.804 sec)
INFO:tensorflow:lr = 0.00015658015 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.4364
INFO:tensorflow:loss = 1.251761, step = 93420 (8.041 sec)
INFO:tensorflow:lr = 0.00015642284 (8.041 sec)
INFO:tensorflow:global_step/sec: 12.9053
INFO:tensorflow:loss = 1.2578083, step = 93520 (7.749 sec)
INFO:tensorflow:lr = 0.00015626573 (7.751 sec)
INFO:tensorflow:global_step/sec: 13.0071
INFO:tensorflow:loss = 1.2468418, step = 93620 (7.687 sec)
INFO:tensorflow:lr = 0.00015610874 (7.686 sec)
INFO:tensorflow:global_step/sec: 12.4707
INFO:tensorflow:loss = 1.2501042, step = 93720 (8.023 sec)
INFO:tensorflow:lr = 0.00015595194 (8.026 sec)
INFO:tensorflow:global_step/sec: 12.5457
INFO:tensorflow:loss = 1.2522388, step = 93820 (7.966 sec)
INFO:tensorflow:lr = 0.00015579528 (7.968 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.2347
INFO:tensorflow:loss = 1.2465913, step = 93920 (10.833 sec)
INFO:tensorflow:lr = 0.00015563879 (10.830 sec)
INFO:tensorflow:global_step/sec: 12.9867
INFO:tensorflow:loss = 1.2459806, step = 94020 (7.702 sec)
INFO:tensorflow:lr = 0.00015548244 (7.702 sec)
INFO:tensorflow:global_step/sec: 12.8375
INFO:tensorflow:loss = 1.2477661, step = 94120 (7.789 sec)
INFO:tensorflow:lr = 0.00015532626 (7.789 sec)
INFO:tensorflow:global_step/sec: 12.7264
INFO:tensorflow:loss = 1.2550383, step = 94220 (7.857 sec)
INFO:tensorflow:lr = 0.00015517023 (7.856 sec)
INFO:tensorflow:global_step/sec: 12.756
INFO:tensorflow:loss = 1.2482506, step = 94320 (7.840 sec)
INFO:tensorflow:lr = 0.00015501435 (7.840 sec)
INFO:tensorflow:global_step/sec: 12.5792
INFO:tensorflow:loss = 1.2513517, step = 94420 (7.949 sec)
INFO:tensorflow:lr = 0.00015485863 (7.951 sec)
INFO:tensorflow:global_step/sec: 12.7862
INFO:tensorflow:loss = 1.2492919, step = 94520 (7.816 sec)
INFO:tensorflow:lr = 0.00015470307 (7.817 sec)
INFO:tensorflow:global_step/sec: 12.8749
INFO:tensorflow:loss = 1.2463424, step = 94620 (7.773 sec)
INFO:tensorflow:lr = 0.00015454767 (7.771 sec)
INFO:tensorflow:global_step/sec: 12.5792
INFO:tensorflow:loss = 1.245236, step = 94720 (7.944 sec)
INFO:tensorflow:lr = 0.00015439242 (7.944 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.961
INFO:tensorflow:loss = 1.2500908, step = 94820 (8.375 sec)
INFO:tensorflow:lr = 0.00015423732 (8.375 sec)
INFO:tensorflow:global_step/sec: 10.729
INFO:tensorflow:loss = 1.2478384, step = 94920 (9.307 sec)
INFO:tensorflow:lr = 0.00015408237 (9.309 sec)
INFO:tensorflow:global_step/sec: 12.6397
INFO:tensorflow:loss = 1.248349, step = 95020 (7.911 sec)
INFO:tensorflow:lr = 0.00015392763 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.7204
INFO:tensorflow:loss = 1.2496285, step = 95120 (7.862 sec)
INFO:tensorflow:lr = 0.000153773 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.7468
INFO:tensorflow:loss = 1.2597445, step = 95220 (7.848 sec)
INFO:tensorflow:lr = 0.00015361853 (7.846 sec)
INFO:tensorflow:global_step/sec: 12.5758
INFO:tensorflow:loss = 1.2453961, step = 95320 (7.952 sec)
INFO:tensorflow:lr = 0.0001534642 (7.950 sec)
INFO:tensorflow:global_step/sec: 12.732
INFO:tensorflow:loss = 1.2543547, step = 95420 (7.852 sec)
INFO:tensorflow:lr = 0.00015331004 (7.852 sec)
INFO:tensorflow:global_step/sec: 12.5471
INFO:tensorflow:loss = 1.2571616, step = 95520 (7.971 sec)
INFO:tensorflow:lr = 0.00015315604 (7.972 sec)
INFO:tensorflow:global_step/sec: 12.8873
INFO:tensorflow:loss = 1.2492933, step = 95620 (7.761 sec)
INFO:tensorflow:lr = 0.0001530022 (7.759 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.9774
INFO:tensorflow:loss = 1.2539256, step = 95720 (7.708 sec)
INFO:tensorflow:lr = 0.0001528485 (7.709 sec)
INFO:tensorflow:global_step/sec: 10.8335
INFO:tensorflow:loss = 1.2586083, step = 95820 (9.226 sec)
INFO:tensorflow:lr = 0.00015269495 (9.226 sec)
INFO:tensorflow:global_step/sec: 12.9627
INFO:tensorflow:loss = 1.2478311, step = 95920 (7.717 sec)
INFO:tensorflow:lr = 0.00015254157 (7.716 sec)
INFO:tensorflow:global_step/sec: 12.7404
INFO:tensorflow:loss = 1.246582, step = 96020 (7.847 sec)
INFO:tensorflow:lr = 0.00015238834 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.8753
INFO:tensorflow:loss = 1.2462858, step = 96120 (7.763 sec)
INFO:tensorflow:lr = 0.00015223527 (7.763 sec)
INFO:tensorflow:global_step/sec: 12.7781
INFO:tensorflow:loss = 1.248619, step = 96220 (7.827 sec)
INFO:tensorflow:lr = 0.00015208234 (7.826 sec)
INFO:tensorflow:global_step/sec: 12.8999
INFO:tensorflow:loss = 1.2437112, step = 96320 (7.751 sec)
INFO:tensorflow:lr = 0.00015192956 (7.751 sec)
INFO:tensorflow:global_step/sec: 12.9556
INFO:tensorflow:loss = 1.2472332, step = 96420 (7.724 sec)
INFO:tensorflow:lr = 0.00015177694 (7.724 sec)
INFO:tensorflow:global_step/sec: 12.5363
INFO:tensorflow:loss = 1.2497894, step = 96520 (7.972 sec)
INFO:tensorflow:lr = 0.00015162448 (7.972 sec)
INFO:tensorflow:global_step/sec: 12.7175
INFO:tensorflow:loss = 1.2479632, step = 96620 (7.868 sec)
INFO:tensorflow:lr = 0.00015147217 (7.867 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.8365
INFO:tensorflow:loss = 1.2539418, step = 96720 (8.445 sec)
INFO:tensorflow:lr = 0.00015132003 (8.448 sec)
INFO:tensorflow:global_step/sec: 11.3935
INFO:tensorflow:loss = 1.246201, step = 96820 (8.780 sec)
INFO:tensorflow:lr = 0.000151168 (8.777 sec)
INFO:tensorflow:global_step/sec: 12.8646
INFO:tensorflow:loss = 1.2464495, step = 96920 (7.773 sec)
INFO:tensorflow:lr = 0.00015101615 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.8559
INFO:tensorflow:loss = 1.2489393, step = 97020 (7.776 sec)
INFO:tensorflow:lr = 0.00015086446 (7.777 sec)
INFO:tensorflow:global_step/sec: 12.755
INFO:tensorflow:loss = 1.2475141, step = 97120 (7.844 sec)
INFO:tensorflow:lr = 0.00015071292 (7.843 sec)
INFO:tensorflow:global_step/sec: 12.469
INFO:tensorflow:loss = 1.2522881, step = 97220 (8.019 sec)
INFO:tensorflow:lr = 0.00015056152 (8.019 sec)
INFO:tensorflow:global_step/sec: 12.8521
INFO:tensorflow:loss = 1.2474856, step = 97320 (7.775 sec)
INFO:tensorflow:lr = 0.00015041027 (7.777 sec)
INFO:tensorflow:global_step/sec: 13.0048
INFO:tensorflow:loss = 1.2518626, step = 97420 (7.691 sec)
INFO:tensorflow:lr = 0.00015025918 (7.692 sec)
INFO:tensorflow:global_step/sec: 12.8331
INFO:tensorflow:loss = 1.2469357, step = 97520 (7.797 sec)
INFO:tensorflow:lr = 0.00015010824 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.8551
INFO:tensorflow:loss = 1.2467525, step = 97620 (7.774 sec)
INFO:tensorflow:lr = 0.00014995746 (7.774 sec)
INFO:tensorflow:global_step/sec: 12.8103
INFO:tensorflow:loss = 1.2500345, step = 97720 (7.811 sec)
INFO:tensorflow:lr = 0.00014980682 (7.811 sec)
INFO:tensorflow:Saving checkpoints for 97800 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2452809.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-97800
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-97800
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.721
INFO:tensorflow:Best (Exact Match) Accuracy: 0.721
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-97800
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 97800 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2478939, step = 97800
INFO:tensorflow:lr = 0.00014968641
INFO:tensorflow:global_step/sec: 10.2999
INFO:tensorflow:loss = 1.2530776, step = 97900 (9.710 sec)
INFO:tensorflow:lr = 0.00014953606 (9.710 sec)
INFO:tensorflow:global_step/sec: 12.5138
INFO:tensorflow:loss = 1.2495674, step = 98000 (7.999 sec)
INFO:tensorflow:lr = 0.00014938583 (7.998 sec)
INFO:tensorflow:global_step/sec: 12.6598
INFO:tensorflow:loss = 1.2456689, step = 98100 (7.892 sec)
INFO:tensorflow:lr = 0.00014923578 (7.893 sec)
INFO:tensorflow:global_step/sec: 12.7983
INFO:tensorflow:loss = 1.2530382, step = 98200 (7.815 sec)
INFO:tensorflow:lr = 0.00014908587 (7.814 sec)
INFO:tensorflow:global_step/sec: 12.7198
INFO:tensorflow:loss = 1.2474073, step = 98300 (7.867 sec)
INFO:tensorflow:lr = 0.00014893612 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.7548
INFO:tensorflow:loss = 1.2511063, step = 98400 (7.839 sec)
INFO:tensorflow:lr = 0.00014878651 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.6978
INFO:tensorflow:loss = 1.2448481, step = 98500 (7.877 sec)
INFO:tensorflow:lr = 0.00014863703 (7.878 sec)
INFO:tensorflow:global_step/sec: 12.9897
INFO:tensorflow:loss = 1.2509502, step = 98600 (7.692 sec)
INFO:tensorflow:lr = 0.00014848771 (7.690 sec)
INFO:tensorflow:global_step/sec: 12.8457
INFO:tensorflow:loss = 1.2510892, step = 98700 (7.785 sec)
INFO:tensorflow:lr = 0.00014833857 (7.785 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.35826
INFO:tensorflow:loss = 1.2454256, step = 98800 (11.964 sec)
INFO:tensorflow:lr = 0.00014818956 (11.964 sec)
INFO:tensorflow:global_step/sec: 12.6956
INFO:tensorflow:loss = 1.2526236, step = 98900 (7.882 sec)
INFO:tensorflow:lr = 0.0001480407 (7.881 sec)
INFO:tensorflow:global_step/sec: 12.6931
INFO:tensorflow:loss = 1.2457856, step = 99000 (7.879 sec)
INFO:tensorflow:lr = 0.00014789199 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.7124
INFO:tensorflow:loss = 1.2481877, step = 99100 (7.865 sec)
INFO:tensorflow:lr = 0.00014774341 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.9415
INFO:tensorflow:loss = 1.2529738, step = 99200 (7.725 sec)
INFO:tensorflow:lr = 0.00014759501 (7.728 sec)
INFO:tensorflow:global_step/sec: 12.51
INFO:tensorflow:loss = 1.2549247, step = 99300 (7.996 sec)
INFO:tensorflow:lr = 0.00014744674 (7.993 sec)
INFO:tensorflow:global_step/sec: 12.8669
INFO:tensorflow:loss = 1.2506918, step = 99400 (7.773 sec)
INFO:tensorflow:lr = 0.00014729865 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.9446
INFO:tensorflow:loss = 1.251098, step = 99500 (7.724 sec)
INFO:tensorflow:lr = 0.00014715067 (7.724 sec)
INFO:tensorflow:global_step/sec: 12.9673
INFO:tensorflow:loss = 1.2509713, step = 99600 (7.708 sec)
INFO:tensorflow:lr = 0.00014700285 (7.713 sec)
INFO:tensorflow:global_step/sec: 12.8537
INFO:tensorflow:loss = 1.2475537, step = 99700 (7.779 sec)
INFO:tensorflow:lr = 0.0001468552 (7.774 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.30087
INFO:tensorflow:loss = 1.249899, step = 99800 (12.051 sec)
INFO:tensorflow:lr = 0.00014670768 (12.050 sec)
INFO:tensorflow:global_step/sec: 12.6009
INFO:tensorflow:loss = 1.2500576, step = 99900 (7.939 sec)
INFO:tensorflow:lr = 0.00014656028 (7.939 sec)
INFO:tensorflow:global_step/sec: 12.6469
INFO:tensorflow:loss = 1.2471527, step = 100000 (7.902 sec)
INFO:tensorflow:lr = 0.00014641306 (7.905 sec)
INFO:tensorflow:global_step/sec: 12.8393
INFO:tensorflow:loss = 1.2491647, step = 100100 (7.791 sec)
INFO:tensorflow:lr = 0.00014626599 (7.788 sec)
INFO:tensorflow:global_step/sec: 12.4661
INFO:tensorflow:loss = 1.2583389, step = 100200 (8.018 sec)
INFO:tensorflow:lr = 0.00014611905 (8.018 sec)
INFO:tensorflow:global_step/sec: 12.7297
INFO:tensorflow:loss = 1.2637594, step = 100300 (7.856 sec)
INFO:tensorflow:lr = 0.00014597227 (7.856 sec)
INFO:tensorflow:global_step/sec: 13.0237
INFO:tensorflow:loss = 1.2442552, step = 100400 (7.679 sec)
INFO:tensorflow:lr = 0.00014582566 (7.679 sec)
INFO:tensorflow:global_step/sec: 13.0379
INFO:tensorflow:loss = 1.2445681, step = 100500 (7.670 sec)
INFO:tensorflow:lr = 0.00014567917 (7.671 sec)
INFO:tensorflow:global_step/sec: 12.9349
INFO:tensorflow:loss = 1.2465498, step = 100600 (7.731 sec)
INFO:tensorflow:lr = 0.00014553282 (7.730 sec)
INFO:tensorflow:global_step/sec: 12.8189
INFO:tensorflow:loss = 1.250596, step = 100700 (7.803 sec)
INFO:tensorflow:lr = 0.00014538663 (7.803 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.28172
INFO:tensorflow:loss = 1.2455262, step = 100800 (12.075 sec)
INFO:tensorflow:lr = 0.0001452406 (12.075 sec)
INFO:tensorflow:global_step/sec: 12.9382
INFO:tensorflow:loss = 1.2536768, step = 100900 (7.732 sec)
INFO:tensorflow:lr = 0.0001450947 (7.732 sec)
INFO:tensorflow:global_step/sec: 12.7612
INFO:tensorflow:loss = 1.2488126, step = 101000 (7.836 sec)
INFO:tensorflow:lr = 0.00014494894 (7.835 sec)
INFO:tensorflow:global_step/sec: 12.9549
INFO:tensorflow:loss = 1.2458501, step = 101100 (7.714 sec)
INFO:tensorflow:lr = 0.00014480333 (7.716 sec)
INFO:tensorflow:global_step/sec: 12.6402
INFO:tensorflow:loss = 1.2484591, step = 101200 (7.912 sec)
INFO:tensorflow:lr = 0.00014465788 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.4116
INFO:tensorflow:loss = 1.2440362, step = 101300 (8.059 sec)
INFO:tensorflow:lr = 0.00014451255 (8.059 sec)
INFO:tensorflow:global_step/sec: 12.6824
INFO:tensorflow:loss = 1.2500799, step = 101400 (7.882 sec)
INFO:tensorflow:lr = 0.00014436738 (7.883 sec)
INFO:tensorflow:global_step/sec: 12.6559
INFO:tensorflow:loss = 1.2466154, step = 101500 (7.901 sec)
INFO:tensorflow:lr = 0.00014422236 (7.900 sec)
INFO:tensorflow:global_step/sec: 13.0559
INFO:tensorflow:loss = 1.2482994, step = 101600 (7.660 sec)
INFO:tensorflow:lr = 0.0001440775 (7.660 sec)
INFO:tensorflow:global_step/sec: 12.8514
INFO:tensorflow:loss = 1.2523829, step = 101700 (7.785 sec)
INFO:tensorflow:lr = 0.00014393278 (7.784 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.45087
INFO:tensorflow:loss = 1.2452083, step = 101800 (11.835 sec)
INFO:tensorflow:lr = 0.00014378819 (11.835 sec)
INFO:tensorflow:global_step/sec: 12.569
INFO:tensorflow:loss = 1.2493188, step = 101900 (7.951 sec)
INFO:tensorflow:lr = 0.00014364375 (7.951 sec)
INFO:tensorflow:global_step/sec: 12.8148
INFO:tensorflow:loss = 1.2473443, step = 102000 (7.804 sec)
INFO:tensorflow:lr = 0.00014349945 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.4523
INFO:tensorflow:loss = 1.2531881, step = 102100 (8.030 sec)
INFO:tensorflow:lr = 0.00014335531 (8.030 sec)
INFO:tensorflow:global_step/sec: 12.6382
INFO:tensorflow:loss = 1.2482383, step = 102200 (7.913 sec)
INFO:tensorflow:lr = 0.0001432113 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.8723
INFO:tensorflow:loss = 1.2539681, step = 102300 (7.768 sec)
INFO:tensorflow:lr = 0.00014306743 (7.771 sec)
INFO:tensorflow:global_step/sec: 12.9976
INFO:tensorflow:loss = 1.2506423, step = 102400 (7.699 sec)
INFO:tensorflow:lr = 0.00014292373 (7.697 sec)
INFO:tensorflow:global_step/sec: 12.6729
INFO:tensorflow:loss = 1.2471061, step = 102500 (7.890 sec)
INFO:tensorflow:lr = 0.00014278015 (7.889 sec)
INFO:tensorflow:global_step/sec: 13.0785
INFO:tensorflow:loss = 1.253192, step = 102600 (7.642 sec)
INFO:tensorflow:lr = 0.00014263671 (7.642 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.65168
INFO:tensorflow:loss = 1.2455392, step = 102700 (11.565 sec)
INFO:tensorflow:lr = 0.00014249345 (11.565 sec)
INFO:tensorflow:global_step/sec: 12.6778
INFO:tensorflow:loss = 1.2455585, step = 102800 (7.882 sec)
INFO:tensorflow:lr = 0.0001423503 (7.882 sec)
INFO:tensorflow:global_step/sec: 12.9287
INFO:tensorflow:loss = 1.2474979, step = 102900 (7.735 sec)
INFO:tensorflow:lr = 0.00014220731 (7.736 sec)
INFO:tensorflow:global_step/sec: 12.5712
INFO:tensorflow:loss = 1.2525347, step = 103000 (7.958 sec)
INFO:tensorflow:lr = 0.00014206445 (7.957 sec)
INFO:tensorflow:global_step/sec: 12.7429
INFO:tensorflow:loss = 1.2527738, step = 103100 (7.848 sec)
INFO:tensorflow:lr = 0.00014192174 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.8437
INFO:tensorflow:loss = 1.2513976, step = 103200 (7.782 sec)
INFO:tensorflow:lr = 0.00014177918 (7.782 sec)
INFO:tensorflow:global_step/sec: 13.1909
INFO:tensorflow:loss = 1.2616798, step = 103300 (7.585 sec)
INFO:tensorflow:lr = 0.00014163676 (7.585 sec)
INFO:tensorflow:global_step/sec: 12.7898
INFO:tensorflow:loss = 1.2459317, step = 103400 (7.819 sec)
INFO:tensorflow:lr = 0.00014149447 (7.817 sec)
INFO:tensorflow:global_step/sec: 12.7356
INFO:tensorflow:loss = 1.2490375, step = 103500 (7.847 sec)
INFO:tensorflow:lr = 0.00014135234 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.6504
INFO:tensorflow:loss = 1.2508307, step = 103600 (7.908 sec)
INFO:tensorflow:lr = 0.00014121036 (7.907 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.35231
INFO:tensorflow:loss = 1.2476784, step = 103700 (10.694 sec)
INFO:tensorflow:lr = 0.00014106852 (10.697 sec)
INFO:tensorflow:global_step/sec: 12.7486
INFO:tensorflow:loss = 1.2524669, step = 103800 (7.845 sec)
INFO:tensorflow:lr = 0.0001409268 (7.843 sec)
INFO:tensorflow:global_step/sec: 12.73
INFO:tensorflow:loss = 1.247279, step = 103900 (7.855 sec)
INFO:tensorflow:lr = 0.00014078524 (7.856 sec)
INFO:tensorflow:global_step/sec: 12.6726
INFO:tensorflow:loss = 1.2515517, step = 104000 (7.890 sec)
INFO:tensorflow:lr = 0.00014064382 (7.890 sec)
INFO:tensorflow:global_step/sec: 12.5949
INFO:tensorflow:loss = 1.243661, step = 104100 (7.935 sec)
INFO:tensorflow:lr = 0.00014050253 (7.935 sec)
INFO:tensorflow:global_step/sec: 12.8437
INFO:tensorflow:loss = 1.2481575, step = 104200 (7.786 sec)
INFO:tensorflow:lr = 0.00014036139 (7.786 sec)
INFO:tensorflow:global_step/sec: 12.7226
INFO:tensorflow:loss = 1.2454779, step = 104300 (7.866 sec)
INFO:tensorflow:lr = 0.00014022039 (7.866 sec)
INFO:tensorflow:global_step/sec: 13.1483
INFO:tensorflow:loss = 1.2469022, step = 104400 (7.601 sec)
INFO:tensorflow:lr = 0.00014007956 (7.603 sec)
INFO:tensorflow:global_step/sec: 12.7578
INFO:tensorflow:loss = 1.2458844, step = 104500 (7.843 sec)
INFO:tensorflow:lr = 0.00013993881 (7.841 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0438
INFO:tensorflow:loss = 1.2457703, step = 104600 (8.302 sec)
INFO:tensorflow:lr = 0.00013979826 (8.302 sec)
INFO:tensorflow:global_step/sec: 10.6714
INFO:tensorflow:loss = 1.2621123, step = 104700 (9.368 sec)
INFO:tensorflow:lr = 0.0001396578 (9.369 sec)
INFO:tensorflow:global_step/sec: 12.7622
INFO:tensorflow:loss = 1.252069, step = 104800 (7.839 sec)
INFO:tensorflow:lr = 0.00013951753 (7.838 sec)
INFO:tensorflow:global_step/sec: 12.8765
INFO:tensorflow:loss = 1.2471956, step = 104900 (7.769 sec)
INFO:tensorflow:lr = 0.00013937739 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.6423
INFO:tensorflow:loss = 1.2450892, step = 105000 (7.908 sec)
INFO:tensorflow:lr = 0.00013923737 (7.909 sec)
INFO:tensorflow:global_step/sec: 12.8286
INFO:tensorflow:loss = 1.2544929, step = 105100 (7.793 sec)
INFO:tensorflow:lr = 0.0001390975 (7.793 sec)
INFO:tensorflow:global_step/sec: 12.5893
INFO:tensorflow:loss = 1.247633, step = 105200 (7.940 sec)
INFO:tensorflow:lr = 0.00013895777 (7.939 sec)
INFO:tensorflow:global_step/sec: 13.1147
INFO:tensorflow:loss = 1.2575452, step = 105300 (7.630 sec)
INFO:tensorflow:lr = 0.00013881818 (7.630 sec)
INFO:tensorflow:global_step/sec: 12.8288
INFO:tensorflow:loss = 1.2458938, step = 105400 (7.789 sec)
INFO:tensorflow:lr = 0.00013867875 (7.792 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.4148
INFO:tensorflow:loss = 1.2475997, step = 105500 (8.062 sec)
INFO:tensorflow:lr = 0.00013853943 (8.060 sec)
INFO:tensorflow:global_step/sec: 11.1919
INFO:tensorflow:loss = 1.2466756, step = 105600 (8.932 sec)
INFO:tensorflow:lr = 0.00013840027 (8.931 sec)
INFO:tensorflow:global_step/sec: 13.0127
INFO:tensorflow:loss = 1.2510452, step = 105700 (7.686 sec)
INFO:tensorflow:lr = 0.00013826125 (7.686 sec)
INFO:tensorflow:global_step/sec: 12.7357
INFO:tensorflow:loss = 1.2437646, step = 105800 (7.850 sec)
INFO:tensorflow:lr = 0.00013812236 (7.849 sec)
INFO:tensorflow:global_step/sec: 12.5414
INFO:tensorflow:loss = 1.2473345, step = 105900 (7.970 sec)
INFO:tensorflow:lr = 0.0001379836 (7.972 sec)
INFO:tensorflow:global_step/sec: 13.1739
INFO:tensorflow:loss = 1.2456942, step = 106000 (7.591 sec)
INFO:tensorflow:lr = 0.000137845 (7.591 sec)
INFO:tensorflow:global_step/sec: 12.8703
INFO:tensorflow:loss = 1.2483459, step = 106100 (7.776 sec)
INFO:tensorflow:lr = 0.00013770652 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.7021
INFO:tensorflow:loss = 1.24482, step = 106200 (7.872 sec)
INFO:tensorflow:lr = 0.00013756819 (7.872 sec)
INFO:tensorflow:global_step/sec: 12.8143
INFO:tensorflow:loss = 1.2478292, step = 106300 (7.801 sec)
INFO:tensorflow:lr = 0.00013743 (7.802 sec)
INFO:tensorflow:global_step/sec: 12.9085
INFO:tensorflow:loss = 1.2532618, step = 106400 (7.749 sec)
INFO:tensorflow:lr = 0.00013729198 (7.749 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.9567
INFO:tensorflow:loss = 1.2503626, step = 106500 (8.363 sec)
INFO:tensorflow:lr = 0.00013715404 (8.363 sec)
INFO:tensorflow:global_step/sec: 11.4986
INFO:tensorflow:loss = 1.2475023, step = 106600 (8.699 sec)
INFO:tensorflow:lr = 0.00013701625 (8.699 sec)
INFO:tensorflow:global_step/sec: 12.6916
INFO:tensorflow:loss = 1.246882, step = 106700 (7.875 sec)
INFO:tensorflow:lr = 0.00013687865 (7.875 sec)
INFO:tensorflow:global_step/sec: 12.9143
INFO:tensorflow:loss = 1.2436101, step = 106800 (7.745 sec)
INFO:tensorflow:lr = 0.00013674115 (7.744 sec)
INFO:tensorflow:global_step/sec: 12.8539
INFO:tensorflow:loss = 1.2468636, step = 106900 (7.779 sec)
INFO:tensorflow:lr = 0.00013660378 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.7943
INFO:tensorflow:loss = 1.2541093, step = 107000 (7.816 sec)
INFO:tensorflow:lr = 0.00013646655 (7.816 sec)
INFO:tensorflow:global_step/sec: 12.9027
INFO:tensorflow:loss = 1.2474351, step = 107100 (7.750 sec)
INFO:tensorflow:lr = 0.00013632947 (7.750 sec)
INFO:tensorflow:global_step/sec: 13.1961
INFO:tensorflow:loss = 1.2451929, step = 107200 (7.574 sec)
INFO:tensorflow:lr = 0.00013619252 (7.578 sec)
INFO:tensorflow:global_step/sec: 12.5269
INFO:tensorflow:loss = 1.2497258, step = 107300 (7.987 sec)
INFO:tensorflow:lr = 0.00013605572 (7.982 sec)
INFO:tensorflow:global_step/sec: 12.5849
INFO:tensorflow:loss = 1.2448314, step = 107400 (7.946 sec)
INFO:tensorflow:lr = 0.00013591905 (7.946 sec)
INFO:tensorflow:global_step/sec: 12.951
INFO:tensorflow:loss = 1.2444179, step = 107500 (7.717 sec)
INFO:tensorflow:lr = 0.00013578251 (7.719 sec)
INFO:tensorflow:Saving checkpoints for 107580 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2442379.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-107580
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-107580
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.723
INFO:tensorflow:Best (Exact Match) Accuracy: 0.723
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-107580
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 107580 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2489259, step = 107580
INFO:tensorflow:lr = 0.00013567338
INFO:tensorflow:global_step/sec: 10.4098
INFO:tensorflow:loss = 1.2465535, step = 107680 (9.612 sec)
INFO:tensorflow:lr = 0.00013553708 (9.612 sec)
INFO:tensorflow:global_step/sec: 12.4733
INFO:tensorflow:loss = 1.249318, step = 107780 (8.013 sec)
INFO:tensorflow:lr = 0.00013540093 (8.012 sec)
INFO:tensorflow:global_step/sec: 12.822
INFO:tensorflow:loss = 1.2475603, step = 107880 (7.800 sec)
INFO:tensorflow:lr = 0.00013526493 (7.802 sec)
INFO:tensorflow:global_step/sec: 13.0186
INFO:tensorflow:loss = 1.2548482, step = 107980 (7.688 sec)
INFO:tensorflow:lr = 0.00013512906 (7.691 sec)
INFO:tensorflow:global_step/sec: 12.6619
INFO:tensorflow:loss = 1.247991, step = 108080 (7.895 sec)
INFO:tensorflow:lr = 0.00013499332 (7.892 sec)
INFO:tensorflow:global_step/sec: 12.4877
INFO:tensorflow:loss = 1.2462689, step = 108180 (8.008 sec)
INFO:tensorflow:lr = 0.00013485769 (8.006 sec)
INFO:tensorflow:global_step/sec: 12.7378
INFO:tensorflow:loss = 1.2458864, step = 108280 (7.845 sec)
INFO:tensorflow:lr = 0.00013472223 (7.846 sec)
INFO:tensorflow:global_step/sec: 12.869
INFO:tensorflow:loss = 1.2465334, step = 108380 (7.783 sec)
INFO:tensorflow:lr = 0.00013458691 (7.784 sec)
INFO:tensorflow:global_step/sec: 12.3673
INFO:tensorflow:loss = 1.2464366, step = 108480 (8.074 sec)
INFO:tensorflow:lr = 0.00013445171 (8.075 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.4047
INFO:tensorflow:loss = 1.2486157, step = 108580 (11.904 sec)
INFO:tensorflow:lr = 0.00013431665 (11.903 sec)
INFO:tensorflow:global_step/sec: 12.9262
INFO:tensorflow:loss = 1.2478218, step = 108680 (7.734 sec)
INFO:tensorflow:lr = 0.00013418171 (7.733 sec)
INFO:tensorflow:global_step/sec: 12.7297
INFO:tensorflow:loss = 1.2497544, step = 108780 (7.853 sec)
INFO:tensorflow:lr = 0.00013404692 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.6782
INFO:tensorflow:loss = 1.2516304, step = 108880 (7.887 sec)
INFO:tensorflow:lr = 0.00013391227 (7.885 sec)
INFO:tensorflow:global_step/sec: 12.8279
INFO:tensorflow:loss = 1.2466173, step = 108980 (7.797 sec)
INFO:tensorflow:lr = 0.00013377776 (7.798 sec)
INFO:tensorflow:global_step/sec: 12.6044
INFO:tensorflow:loss = 1.2438627, step = 109080 (7.936 sec)
INFO:tensorflow:lr = 0.00013364338 (7.937 sec)
INFO:tensorflow:global_step/sec: 12.5006
INFO:tensorflow:loss = 1.2459699, step = 109180 (7.995 sec)
INFO:tensorflow:lr = 0.00013350912 (7.999 sec)
INFO:tensorflow:global_step/sec: 12.3132
INFO:tensorflow:loss = 1.2494336, step = 109280 (8.128 sec)
INFO:tensorflow:lr = 0.00013337501 (8.123 sec)
INFO:tensorflow:global_step/sec: 12.6142
INFO:tensorflow:loss = 1.2490482, step = 109380 (7.923 sec)
INFO:tensorflow:lr = 0.00013324103 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.9434
INFO:tensorflow:loss = 1.244992, step = 109480 (7.731 sec)
INFO:tensorflow:lr = 0.00013310718 (7.727 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.42797
INFO:tensorflow:loss = 1.246099, step = 109580 (11.859 sec)
INFO:tensorflow:lr = 0.00013297348 (11.860 sec)
INFO:tensorflow:global_step/sec: 12.7311
INFO:tensorflow:loss = 1.2468445, step = 109680 (7.856 sec)
INFO:tensorflow:lr = 0.00013283991 (7.855 sec)
INFO:tensorflow:global_step/sec: 12.4444
INFO:tensorflow:loss = 1.2453207, step = 109780 (8.036 sec)
INFO:tensorflow:lr = 0.00013270645 (8.041 sec)
INFO:tensorflow:global_step/sec: 12.396
INFO:tensorflow:loss = 1.247017, step = 109880 (8.070 sec)
INFO:tensorflow:lr = 0.00013257316 (8.066 sec)
INFO:tensorflow:global_step/sec: 12.881
INFO:tensorflow:loss = 1.2435148, step = 109980 (7.759 sec)
INFO:tensorflow:lr = 0.00013244 (7.759 sec)
INFO:tensorflow:global_step/sec: 12.5229
INFO:tensorflow:loss = 1.2592441, step = 110080 (7.990 sec)
INFO:tensorflow:lr = 0.00013230696 (7.990 sec)
INFO:tensorflow:global_step/sec: 12.6508
INFO:tensorflow:loss = 1.2503728, step = 110180 (7.900 sec)
INFO:tensorflow:lr = 0.00013217404 (7.902 sec)
INFO:tensorflow:global_step/sec: 12.2981
INFO:tensorflow:loss = 1.2484401, step = 110280 (8.134 sec)
INFO:tensorflow:lr = 0.00013204126 (8.134 sec)
INFO:tensorflow:global_step/sec: 12.6796
INFO:tensorflow:loss = 1.2463096, step = 110380 (7.885 sec)
INFO:tensorflow:lr = 0.00013190863 (7.886 sec)
INFO:tensorflow:global_step/sec: 12.2949
INFO:tensorflow:loss = 1.2451106, step = 110480 (8.141 sec)
INFO:tensorflow:lr = 0.00013177612 (8.138 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.84506
INFO:tensorflow:loss = 1.2448896, step = 110580 (12.743 sec)
INFO:tensorflow:lr = 0.00013164374 (12.744 sec)
INFO:tensorflow:global_step/sec: 12.412
INFO:tensorflow:loss = 1.2469372, step = 110680 (8.057 sec)
INFO:tensorflow:lr = 0.0001315115 (8.056 sec)
INFO:tensorflow:global_step/sec: 12.3222
INFO:tensorflow:loss = 1.2455852, step = 110780 (8.112 sec)
INFO:tensorflow:lr = 0.0001313794 (8.113 sec)
INFO:tensorflow:global_step/sec: 11.929
INFO:tensorflow:loss = 1.2448218, step = 110880 (8.381 sec)
INFO:tensorflow:lr = 0.00013124742 (8.379 sec)
INFO:tensorflow:global_step/sec: 12.2422
INFO:tensorflow:loss = 1.2461655, step = 110980 (8.168 sec)
INFO:tensorflow:lr = 0.0001311156 (8.168 sec)
INFO:tensorflow:global_step/sec: 12.5866
INFO:tensorflow:loss = 1.2536453, step = 111080 (7.951 sec)
INFO:tensorflow:lr = 0.00013098388 (7.951 sec)
INFO:tensorflow:global_step/sec: 12.4155
INFO:tensorflow:loss = 1.2434506, step = 111180 (8.055 sec)
INFO:tensorflow:lr = 0.00013085229 (8.059 sec)
INFO:tensorflow:global_step/sec: 12.2427
INFO:tensorflow:loss = 1.2464231, step = 111280 (8.167 sec)
INFO:tensorflow:lr = 0.00013072084 (8.166 sec)
INFO:tensorflow:global_step/sec: 12.5907
INFO:tensorflow:loss = 1.2453758, step = 111380 (7.938 sec)
INFO:tensorflow:lr = 0.00013058954 (7.937 sec)
INFO:tensorflow:global_step/sec: 12.716
INFO:tensorflow:loss = 1.2442877, step = 111480 (7.864 sec)
INFO:tensorflow:lr = 0.00013045836 (7.866 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.21248
INFO:tensorflow:loss = 1.2450273, step = 111580 (12.182 sec)
INFO:tensorflow:lr = 0.00013032732 (12.177 sec)
INFO:tensorflow:global_step/sec: 12.3679
INFO:tensorflow:loss = 1.2450981, step = 111680 (8.082 sec)
INFO:tensorflow:lr = 0.0001301964 (8.086 sec)
INFO:tensorflow:global_step/sec: 12.2747
INFO:tensorflow:loss = 1.2465881, step = 111780 (8.144 sec)
INFO:tensorflow:lr = 0.0001300656 (8.141 sec)
INFO:tensorflow:global_step/sec: 12.6075
INFO:tensorflow:loss = 1.2524883, step = 111880 (7.933 sec)
INFO:tensorflow:lr = 0.00012993495 (7.933 sec)
INFO:tensorflow:global_step/sec: 12.1696
INFO:tensorflow:loss = 1.2459798, step = 111980 (8.216 sec)
INFO:tensorflow:lr = 0.00012980442 (8.217 sec)
INFO:tensorflow:global_step/sec: 12.4295
INFO:tensorflow:loss = 1.2500193, step = 112080 (8.049 sec)
INFO:tensorflow:lr = 0.00012967402 (8.049 sec)
INFO:tensorflow:global_step/sec: 12.3575
INFO:tensorflow:loss = 1.2450176, step = 112180 (8.088 sec)
INFO:tensorflow:lr = 0.00012954377 (8.088 sec)
INFO:tensorflow:global_step/sec: 12.4561
INFO:tensorflow:loss = 1.2457551, step = 112280 (8.034 sec)
INFO:tensorflow:lr = 0.00012941365 (8.034 sec)
INFO:tensorflow:global_step/sec: 12.7345
INFO:tensorflow:loss = 1.2456206, step = 112380 (7.853 sec)
INFO:tensorflow:lr = 0.00012928365 (7.853 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.4965
INFO:tensorflow:loss = 1.2451046, step = 112480 (11.770 sec)
INFO:tensorflow:lr = 0.00012915379 (11.775 sec)
INFO:tensorflow:global_step/sec: 12.5032
INFO:tensorflow:loss = 1.2449356, step = 112580 (7.997 sec)
INFO:tensorflow:lr = 0.00012902405 (7.991 sec)
INFO:tensorflow:global_step/sec: 12.4797
INFO:tensorflow:loss = 1.2540485, step = 112680 (8.008 sec)
INFO:tensorflow:lr = 0.00012889442 (8.010 sec)
INFO:tensorflow:global_step/sec: 12.4347
INFO:tensorflow:loss = 1.2504659, step = 112780 (8.048 sec)
INFO:tensorflow:lr = 0.00012876495 (8.047 sec)
INFO:tensorflow:global_step/sec: 12.3055
INFO:tensorflow:loss = 1.2495759, step = 112880 (8.123 sec)
INFO:tensorflow:lr = 0.00012863561 (8.123 sec)
INFO:tensorflow:global_step/sec: 12.5325
INFO:tensorflow:loss = 1.2485863, step = 112980 (7.979 sec)
INFO:tensorflow:lr = 0.00012850638 (7.981 sec)
INFO:tensorflow:global_step/sec: 12.3319
INFO:tensorflow:loss = 1.2465398, step = 113080 (8.107 sec)
INFO:tensorflow:lr = 0.00012837729 (8.112 sec)
INFO:tensorflow:global_step/sec: 12.3829
INFO:tensorflow:loss = 1.2465779, step = 113180 (8.076 sec)
INFO:tensorflow:lr = 0.00012824834 (8.069 sec)
INFO:tensorflow:global_step/sec: 12.6251
INFO:tensorflow:loss = 1.245736, step = 113280 (7.925 sec)
INFO:tensorflow:lr = 0.0001281195 (7.926 sec)
INFO:tensorflow:global_step/sec: 12.3343
INFO:tensorflow:loss = 1.243198, step = 113380 (8.106 sec)
INFO:tensorflow:lr = 0.0001279908 (8.109 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.77301
INFO:tensorflow:loss = 1.2464836, step = 113480 (11.396 sec)
INFO:tensorflow:lr = 0.00012786225 (11.392 sec)
INFO:tensorflow:global_step/sec: 12.4675
INFO:tensorflow:loss = 1.2475723, step = 113580 (8.026 sec)
INFO:tensorflow:lr = 0.0001277338 (8.026 sec)
INFO:tensorflow:global_step/sec: 12.5777
INFO:tensorflow:loss = 1.2461356, step = 113680 (7.945 sec)
INFO:tensorflow:lr = 0.00012760548 (7.945 sec)
INFO:tensorflow:global_step/sec: 12.383
INFO:tensorflow:loss = 1.2514325, step = 113780 (8.077 sec)
INFO:tensorflow:lr = 0.0001274773 (8.080 sec)
INFO:tensorflow:global_step/sec: 12.3492
INFO:tensorflow:loss = 1.2455924, step = 113880 (8.096 sec)
INFO:tensorflow:lr = 0.00012734925 (8.095 sec)
INFO:tensorflow:global_step/sec: 12.3347
INFO:tensorflow:loss = 1.2465392, step = 113980 (8.108 sec)
INFO:tensorflow:lr = 0.00012722133 (8.107 sec)
INFO:tensorflow:global_step/sec: 12.4384
INFO:tensorflow:loss = 1.2566938, step = 114080 (8.044 sec)
INFO:tensorflow:lr = 0.00012709353 (8.043 sec)
INFO:tensorflow:global_step/sec: 12.5175
INFO:tensorflow:loss = 1.2492527, step = 114180 (7.983 sec)
INFO:tensorflow:lr = 0.00012696585 (7.988 sec)
INFO:tensorflow:global_step/sec: 12.4114
INFO:tensorflow:loss = 1.2554555, step = 114280 (8.063 sec)
INFO:tensorflow:lr = 0.00012683832 (8.058 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.6301
INFO:tensorflow:loss = 1.248702, step = 114380 (8.610 sec)
INFO:tensorflow:lr = 0.0001267109 (8.611 sec)
INFO:tensorflow:global_step/sec: 10.0063
INFO:tensorflow:loss = 1.25149, step = 114480 (9.980 sec)
INFO:tensorflow:lr = 0.00012658362 (9.982 sec)
INFO:tensorflow:global_step/sec: 11.686
INFO:tensorflow:loss = 1.2474595, step = 114580 (8.554 sec)
INFO:tensorflow:lr = 0.00012645648 (8.552 sec)
INFO:tensorflow:global_step/sec: 12.569
INFO:tensorflow:loss = 1.2511866, step = 114680 (7.956 sec)
INFO:tensorflow:lr = 0.00012632943 (7.955 sec)
INFO:tensorflow:global_step/sec: 12.6191
INFO:tensorflow:loss = 1.2508888, step = 114780 (7.935 sec)
INFO:tensorflow:lr = 0.00012620253 (7.935 sec)
INFO:tensorflow:global_step/sec: 12.1091
INFO:tensorflow:loss = 1.2474946, step = 114880 (8.257 sec)
INFO:tensorflow:lr = 0.00012607576 (8.256 sec)
INFO:tensorflow:global_step/sec: 12.4553
INFO:tensorflow:loss = 1.2473083, step = 114980 (8.023 sec)
INFO:tensorflow:lr = 0.00012594911 (8.024 sec)
INFO:tensorflow:global_step/sec: 12.4882
INFO:tensorflow:loss = 1.2468796, step = 115080 (8.005 sec)
INFO:tensorflow:lr = 0.0001258226 (8.005 sec)
INFO:tensorflow:global_step/sec: 12.3851
INFO:tensorflow:loss = 1.2520475, step = 115180 (8.079 sec)
INFO:tensorflow:lr = 0.0001256962 (8.079 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2298
INFO:tensorflow:loss = 1.2488706, step = 115280 (8.174 sec)
INFO:tensorflow:lr = 0.00012556993 (8.177 sec)
INFO:tensorflow:global_step/sec: 10.543
INFO:tensorflow:loss = 1.2459799, step = 115380 (9.498 sec)
INFO:tensorflow:lr = 0.0001254438 (9.495 sec)
INFO:tensorflow:global_step/sec: 12.1713
INFO:tensorflow:loss = 1.2485315, step = 115480 (8.202 sec)
INFO:tensorflow:lr = 0.00012531779 (8.202 sec)
INFO:tensorflow:global_step/sec: 12.2389
INFO:tensorflow:loss = 1.2580112, step = 115580 (8.170 sec)
INFO:tensorflow:lr = 0.0001251919 (8.170 sec)
INFO:tensorflow:global_step/sec: 12.1377
INFO:tensorflow:loss = 1.2466756, step = 115680 (8.240 sec)
INFO:tensorflow:lr = 0.00012506612 (8.240 sec)
INFO:tensorflow:global_step/sec: 12.202
INFO:tensorflow:loss = 1.2510239, step = 115780 (8.194 sec)
INFO:tensorflow:lr = 0.00012494052 (8.194 sec)
INFO:tensorflow:global_step/sec: 12.4075
INFO:tensorflow:loss = 1.2495364, step = 115880 (8.066 sec)
INFO:tensorflow:lr = 0.000124815 (8.066 sec)
INFO:tensorflow:global_step/sec: 12.1013
INFO:tensorflow:loss = 1.2435124, step = 115980 (8.257 sec)
INFO:tensorflow:lr = 0.00012468963 (8.258 sec)
INFO:tensorflow:global_step/sec: 12.1832
INFO:tensorflow:loss = 1.2507396, step = 116080 (8.216 sec)
INFO:tensorflow:lr = 0.00012456437 (8.216 sec)
INFO:tensorflow:global_step/sec: 12.5537
INFO:tensorflow:loss = 1.2438142, step = 116180 (7.965 sec)
INFO:tensorflow:lr = 0.00012443922 (7.964 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.6062
INFO:tensorflow:loss = 1.2460523, step = 116280 (8.614 sec)
INFO:tensorflow:lr = 0.00012431422 (8.616 sec)
INFO:tensorflow:global_step/sec: 11.1773
INFO:tensorflow:loss = 1.2480481, step = 116380 (8.947 sec)
INFO:tensorflow:lr = 0.00012418936 (8.946 sec)
INFO:tensorflow:global_step/sec: 12.1484
INFO:tensorflow:loss = 1.2447939, step = 116480 (8.233 sec)
INFO:tensorflow:lr = 0.00012406461 (8.232 sec)
INFO:tensorflow:global_step/sec: 12.3809
INFO:tensorflow:loss = 1.2446187, step = 116580 (8.075 sec)
INFO:tensorflow:lr = 0.00012393999 (8.078 sec)
INFO:tensorflow:global_step/sec: 12.4624
INFO:tensorflow:loss = 1.2526968, step = 116680 (8.023 sec)
INFO:tensorflow:lr = 0.00012381548 (8.021 sec)
INFO:tensorflow:global_step/sec: 12.1608
INFO:tensorflow:loss = 1.2439911, step = 116780 (8.221 sec)
INFO:tensorflow:lr = 0.00012369109 (8.219 sec)
INFO:tensorflow:global_step/sec: 12.694
INFO:tensorflow:loss = 1.2494287, step = 116880 (7.878 sec)
INFO:tensorflow:lr = 0.00012356685 (7.878 sec)
INFO:tensorflow:global_step/sec: 12.4984
INFO:tensorflow:loss = 1.2488147, step = 116980 (8.001 sec)
INFO:tensorflow:lr = 0.00012344273 (8.002 sec)
INFO:tensorflow:global_step/sec: 12.1185
INFO:tensorflow:loss = 1.2587779, step = 117080 (8.259 sec)
INFO:tensorflow:lr = 0.00012331874 (8.259 sec)
INFO:tensorflow:global_step/sec: 12.2015
INFO:tensorflow:loss = 1.2470039, step = 117180 (8.188 sec)
INFO:tensorflow:lr = 0.00012319484 (8.187 sec)
INFO:tensorflow:global_step/sec: 12.411
INFO:tensorflow:loss = 1.2492013, step = 117280 (8.059 sec)
INFO:tensorflow:lr = 0.00012307108 (8.063 sec)
INFO:tensorflow:Saving checkpoints for 117360 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2417642.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-117360
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-117360
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.722
INFO:tensorflow:Best (Exact Match) Accuracy: 0.723
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-117360
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 117360 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2482007, step = 117360
INFO:tensorflow:lr = 0.00012297218
INFO:tensorflow:global_step/sec: 9.57093
INFO:tensorflow:loss = 1.2506714, step = 117460 (10.451 sec)
INFO:tensorflow:lr = 0.00012284865 (10.446 sec)
INFO:tensorflow:global_step/sec: 12.0906
INFO:tensorflow:loss = 1.2449813, step = 117560 (8.271 sec)
INFO:tensorflow:lr = 0.00012272525 (8.270 sec)
INFO:tensorflow:global_step/sec: 12.3201
INFO:tensorflow:loss = 1.2460945, step = 117660 (8.116 sec)
INFO:tensorflow:lr = 0.00012260195 (8.116 sec)
INFO:tensorflow:global_step/sec: 12.323
INFO:tensorflow:loss = 1.2516643, step = 117760 (8.121 sec)
INFO:tensorflow:lr = 0.00012247881 (8.127 sec)
INFO:tensorflow:global_step/sec: 12.4119
INFO:tensorflow:loss = 1.2460521, step = 117860 (8.054 sec)
INFO:tensorflow:lr = 0.00012235578 (8.048 sec)
INFO:tensorflow:global_step/sec: 12.5833
INFO:tensorflow:loss = 1.2443562, step = 117960 (7.950 sec)
INFO:tensorflow:lr = 0.00012223287 (7.955 sec)
INFO:tensorflow:global_step/sec: 12.1709
INFO:tensorflow:loss = 1.246166, step = 118060 (8.216 sec)
INFO:tensorflow:lr = 0.00012211008 (8.211 sec)
INFO:tensorflow:global_step/sec: 12.5068
INFO:tensorflow:loss = 1.2477903, step = 118160 (7.995 sec)
INFO:tensorflow:lr = 0.00012198741 (7.995 sec)
INFO:tensorflow:global_step/sec: 12.5401
INFO:tensorflow:loss = 1.2436001, step = 118260 (7.969 sec)
INFO:tensorflow:lr = 0.000121864876 (7.970 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.95747
INFO:tensorflow:loss = 1.2526491, step = 118360 (12.569 sec)
INFO:tensorflow:lr = 0.00012174245 (12.569 sec)
INFO:tensorflow:global_step/sec: 12.5398
INFO:tensorflow:loss = 1.247338, step = 118460 (7.972 sec)
INFO:tensorflow:lr = 0.00012162015 (7.972 sec)
INFO:tensorflow:global_step/sec: 12.5136
INFO:tensorflow:loss = 1.2485074, step = 118560 (7.995 sec)
INFO:tensorflow:lr = 0.000121498 (7.995 sec)
INFO:tensorflow:global_step/sec: 12.5232
INFO:tensorflow:loss = 1.2500961, step = 118660 (7.983 sec)
INFO:tensorflow:lr = 0.00012137595 (7.984 sec)
INFO:tensorflow:global_step/sec: 12.4087
INFO:tensorflow:loss = 1.2486347, step = 118760 (8.059 sec)
INFO:tensorflow:lr = 0.00012125402 (8.060 sec)
INFO:tensorflow:global_step/sec: 12.2352
INFO:tensorflow:loss = 1.2490938, step = 118860 (8.177 sec)
INFO:tensorflow:lr = 0.000121132216 (8.176 sec)
INFO:tensorflow:global_step/sec: 12.2436
INFO:tensorflow:loss = 1.25471, step = 118960 (8.168 sec)
INFO:tensorflow:lr = 0.00012101054 (8.170 sec)
INFO:tensorflow:global_step/sec: 12.4951
INFO:tensorflow:loss = 1.2441717, step = 119060 (7.996 sec)
INFO:tensorflow:lr = 0.000120888995 (7.996 sec)
INFO:tensorflow:global_step/sec: 12.6037
INFO:tensorflow:loss = 1.2489609, step = 119160 (7.934 sec)
INFO:tensorflow:lr = 0.000120767545 (7.933 sec)
INFO:tensorflow:global_step/sec: 12.475
INFO:tensorflow:loss = 1.2471738, step = 119260 (8.016 sec)
INFO:tensorflow:lr = 0.000120646226 (8.017 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.94323
INFO:tensorflow:loss = 1.2460295, step = 119360 (12.589 sec)
INFO:tensorflow:lr = 0.00012052504 (12.587 sec)
INFO:tensorflow:global_step/sec: 12.5354
INFO:tensorflow:loss = 1.2540466, step = 119460 (7.980 sec)
INFO:tensorflow:lr = 0.00012040395 (7.982 sec)
INFO:tensorflow:global_step/sec: 12.617
INFO:tensorflow:loss = 1.2463863, step = 119560 (7.925 sec)
INFO:tensorflow:lr = 0.00012028302 (7.923 sec)
INFO:tensorflow:global_step/sec: 12.2514
INFO:tensorflow:loss = 1.2426605, step = 119660 (8.162 sec)
INFO:tensorflow:lr = 0.0001201622 (8.167 sec)
INFO:tensorflow:global_step/sec: 12.4442
INFO:tensorflow:loss = 1.2469814, step = 119760 (8.039 sec)
INFO:tensorflow:lr = 0.000120041484 (8.034 sec)
INFO:tensorflow:global_step/sec: 12.5949
INFO:tensorflow:loss = 1.2479645, step = 119860 (7.940 sec)
INFO:tensorflow:lr = 0.00011992089 (7.940 sec)
INFO:tensorflow:global_step/sec: 12.3139
INFO:tensorflow:loss = 1.2518939, step = 119960 (8.116 sec)
INFO:tensorflow:lr = 0.00011980043 (8.117 sec)
INFO:tensorflow:global_step/sec: 12.3963
INFO:tensorflow:loss = 1.2455319, step = 120060 (8.067 sec)
INFO:tensorflow:lr = 0.000119680095 (8.066 sec)
INFO:tensorflow:global_step/sec: 12.3853
INFO:tensorflow:loss = 1.2456839, step = 120160 (8.075 sec)
INFO:tensorflow:lr = 0.00011955988 (8.075 sec)
INFO:tensorflow:global_step/sec: 12.1621
INFO:tensorflow:loss = 1.2468312, step = 120260 (8.227 sec)
INFO:tensorflow:lr = 0.00011943976 (8.226 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.16586
INFO:tensorflow:loss = 1.2442948, step = 120360 (12.246 sec)
INFO:tensorflow:lr = 0.0001193198 (12.246 sec)
INFO:tensorflow:global_step/sec: 12.5876
INFO:tensorflow:loss = 1.2436161, step = 120460 (7.939 sec)
INFO:tensorflow:lr = 0.00011919992 (7.940 sec)
INFO:tensorflow:global_step/sec: 12.3838
INFO:tensorflow:loss = 1.2517046, step = 120560 (8.075 sec)
INFO:tensorflow:lr = 0.000119080185 (8.075 sec)
INFO:tensorflow:global_step/sec: 12.4207
INFO:tensorflow:loss = 1.2490052, step = 120660 (8.051 sec)
INFO:tensorflow:lr = 0.00011896058 (8.055 sec)
INFO:tensorflow:global_step/sec: 12.6772
INFO:tensorflow:loss = 1.2606004, step = 120760 (7.894 sec)
INFO:tensorflow:lr = 0.00011884108 (7.890 sec)
INFO:tensorflow:global_step/sec: 12.4873
INFO:tensorflow:loss = 1.2480586, step = 120860 (8.002 sec)
INFO:tensorflow:lr = 0.0001187217 (8.008 sec)
INFO:tensorflow:global_step/sec: 12.4646
INFO:tensorflow:loss = 1.2461877, step = 120960 (8.022 sec)
INFO:tensorflow:lr = 0.00011860243 (8.019 sec)
INFO:tensorflow:global_step/sec: 12.6909
INFO:tensorflow:loss = 1.2459522, step = 121060 (7.885 sec)
INFO:tensorflow:lr = 0.00011848329 (7.882 sec)
INFO:tensorflow:global_step/sec: 12.7884
INFO:tensorflow:loss = 1.2450316, step = 121160 (7.819 sec)
INFO:tensorflow:lr = 0.000118364274 (7.820 sec)
INFO:tensorflow:global_step/sec: 12.5167
INFO:tensorflow:loss = 1.2498332, step = 121260 (7.989 sec)
INFO:tensorflow:lr = 0.00011824537 (7.990 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.42824
INFO:tensorflow:loss = 1.2524766, step = 121360 (11.862 sec)
INFO:tensorflow:lr = 0.00011812659 (11.861 sec)
INFO:tensorflow:global_step/sec: 12.8493
INFO:tensorflow:loss = 1.2486544, step = 121460 (7.781 sec)
INFO:tensorflow:lr = 0.000118007934 (7.782 sec)
INFO:tensorflow:global_step/sec: 12.5805
INFO:tensorflow:loss = 1.2440406, step = 121560 (7.949 sec)
INFO:tensorflow:lr = 0.00011788939 (7.954 sec)
INFO:tensorflow:global_step/sec: 12.2106
INFO:tensorflow:loss = 1.24541, step = 121660 (8.191 sec)
INFO:tensorflow:lr = 0.000117770964 (8.185 sec)
INFO:tensorflow:global_step/sec: 12.734
INFO:tensorflow:loss = 1.244577, step = 121760 (7.858 sec)
INFO:tensorflow:lr = 0.00011765268 (7.857 sec)
INFO:tensorflow:global_step/sec: 12.524
INFO:tensorflow:loss = 1.247097, step = 121860 (7.980 sec)
INFO:tensorflow:lr = 0.000117534466 (7.985 sec)
INFO:tensorflow:global_step/sec: 12.346
INFO:tensorflow:loss = 1.2459865, step = 121960 (8.101 sec)
INFO:tensorflow:lr = 0.000117416406 (8.095 sec)
INFO:tensorflow:global_step/sec: 12.6204
INFO:tensorflow:loss = 1.2476454, step = 122060 (7.929 sec)
INFO:tensorflow:lr = 0.000117298456 (7.933 sec)
INFO:tensorflow:global_step/sec: 12.7997
INFO:tensorflow:loss = 1.2479577, step = 122160 (7.806 sec)
INFO:tensorflow:lr = 0.00011718063 (7.803 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.75356
INFO:tensorflow:loss = 1.2421271, step = 122260 (11.424 sec)
INFO:tensorflow:lr = 0.000117062926 (11.427 sec)
INFO:tensorflow:global_step/sec: 12.6182
INFO:tensorflow:loss = 1.2467126, step = 122360 (7.931 sec)
INFO:tensorflow:lr = 0.000116945324 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.2161
INFO:tensorflow:loss = 1.2463832, step = 122460 (8.180 sec)
INFO:tensorflow:lr = 0.00011682784 (8.180 sec)
INFO:tensorflow:global_step/sec: 12.5557
INFO:tensorflow:loss = 1.2444496, step = 122560 (7.965 sec)
INFO:tensorflow:lr = 0.00011671051 (7.966 sec)
INFO:tensorflow:global_step/sec: 12.56
INFO:tensorflow:loss = 1.2485607, step = 122660 (7.968 sec)
INFO:tensorflow:lr = 0.00011659325 (7.971 sec)
INFO:tensorflow:global_step/sec: 12.5083
INFO:tensorflow:loss = 1.2494316, step = 122760 (7.990 sec)
INFO:tensorflow:lr = 0.00011647614 (7.986 sec)
INFO:tensorflow:global_step/sec: 12.8594
INFO:tensorflow:loss = 1.2474501, step = 122860 (7.781 sec)
INFO:tensorflow:lr = 0.00011635912 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.7281
INFO:tensorflow:loss = 1.2478578, step = 122960 (7.856 sec)
INFO:tensorflow:lr = 0.00011624224 (7.857 sec)
INFO:tensorflow:global_step/sec: 12.8408
INFO:tensorflow:loss = 1.2520856, step = 123060 (7.783 sec)
INFO:tensorflow:lr = 0.000116125484 (7.782 sec)
INFO:tensorflow:global_step/sec: 12.8542
INFO:tensorflow:loss = 1.2521843, step = 123160 (7.782 sec)
INFO:tensorflow:lr = 0.00011600883 (7.782 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.06857
INFO:tensorflow:loss = 1.2535832, step = 123260 (11.027 sec)
INFO:tensorflow:lr = 0.0001158923 (11.027 sec)
INFO:tensorflow:global_step/sec: 12.9486
INFO:tensorflow:loss = 1.249761, step = 123360 (7.720 sec)
INFO:tensorflow:lr = 0.00011577587 (7.720 sec)
INFO:tensorflow:global_step/sec: 12.4898
INFO:tensorflow:loss = 1.2453451, step = 123460 (8.013 sec)
INFO:tensorflow:lr = 0.00011565958 (8.013 sec)
INFO:tensorflow:global_step/sec: 12.6435
INFO:tensorflow:loss = 1.2508979, step = 123560 (7.908 sec)
INFO:tensorflow:lr = 0.0001155434 (7.908 sec)
INFO:tensorflow:global_step/sec: 12.5327
INFO:tensorflow:loss = 1.24587, step = 123660 (7.974 sec)
INFO:tensorflow:lr = 0.000115427334 (7.975 sec)
INFO:tensorflow:global_step/sec: 12.9855
INFO:tensorflow:loss = 1.2542604, step = 123760 (7.701 sec)
INFO:tensorflow:lr = 0.000115311384 (7.701 sec)
INFO:tensorflow:global_step/sec: 12.8748
INFO:tensorflow:loss = 1.244994, step = 123860 (7.767 sec)
INFO:tensorflow:lr = 0.000115195544 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.2358
INFO:tensorflow:loss = 1.2472664, step = 123960 (8.173 sec)
INFO:tensorflow:lr = 0.00011507982 (8.172 sec)
INFO:tensorflow:global_step/sec: 12.6827
INFO:tensorflow:loss = 1.2551069, step = 124060 (7.890 sec)
INFO:tensorflow:lr = 0.000114964234 (7.888 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0551
INFO:tensorflow:loss = 1.2484986, step = 124160 (8.296 sec)
INFO:tensorflow:lr = 0.00011484874 (8.297 sec)
INFO:tensorflow:global_step/sec: 10.5292
INFO:tensorflow:loss = 1.2491708, step = 124260 (9.497 sec)
INFO:tensorflow:lr = 0.00011473338 (9.496 sec)
INFO:tensorflow:global_step/sec: 12.902
INFO:tensorflow:loss = 1.2463233, step = 124360 (7.745 sec)
INFO:tensorflow:lr = 0.00011461812 (7.746 sec)
INFO:tensorflow:global_step/sec: 12.766
INFO:tensorflow:loss = 1.2511879, step = 124460 (7.837 sec)
INFO:tensorflow:lr = 0.000114502975 (7.838 sec)
INFO:tensorflow:global_step/sec: 12.7431
INFO:tensorflow:loss = 1.2464348, step = 124560 (7.845 sec)
INFO:tensorflow:lr = 0.00011438796 (7.845 sec)
INFO:tensorflow:global_step/sec: 12.3308
INFO:tensorflow:loss = 1.2550204, step = 124660 (8.114 sec)
INFO:tensorflow:lr = 0.000114273054 (8.113 sec)
INFO:tensorflow:global_step/sec: 12.6927
INFO:tensorflow:loss = 1.2458317, step = 124760 (7.879 sec)
INFO:tensorflow:lr = 0.00011415826 (7.880 sec)
INFO:tensorflow:global_step/sec: 12.6561
INFO:tensorflow:loss = 1.2493947, step = 124860 (7.900 sec)
INFO:tensorflow:lr = 0.00011404359 (7.900 sec)
INFO:tensorflow:global_step/sec: 12.6385
INFO:tensorflow:loss = 1.2473803, step = 124960 (7.912 sec)
INFO:tensorflow:lr = 0.00011392903 (7.912 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2313
INFO:tensorflow:loss = 1.2489656, step = 125060 (8.177 sec)
INFO:tensorflow:lr = 0.000113814574 (8.181 sec)
INFO:tensorflow:global_step/sec: 10.8729
INFO:tensorflow:loss = 1.2458106, step = 125160 (9.195 sec)
INFO:tensorflow:lr = 0.00011370025 (9.192 sec)
INFO:tensorflow:global_step/sec: 12.9656
INFO:tensorflow:loss = 1.245927, step = 125260 (7.709 sec)
INFO:tensorflow:lr = 0.000113586044 (7.712 sec)
INFO:tensorflow:global_step/sec: 12.8824
INFO:tensorflow:loss = 1.245289, step = 125360 (7.763 sec)
INFO:tensorflow:lr = 0.000113471935 (7.760 sec)
INFO:tensorflow:global_step/sec: 12.6135
INFO:tensorflow:loss = 1.244993, step = 125460 (7.928 sec)
INFO:tensorflow:lr = 0.00011335795 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.6983
INFO:tensorflow:loss = 1.246616, step = 125560 (7.875 sec)
INFO:tensorflow:lr = 0.00011324408 (7.875 sec)
INFO:tensorflow:global_step/sec: 12.7768
INFO:tensorflow:loss = 1.243581, step = 125660 (7.830 sec)
INFO:tensorflow:lr = 0.000113130314 (7.830 sec)
INFO:tensorflow:global_step/sec: 12.5408
INFO:tensorflow:loss = 1.2462535, step = 125760 (7.975 sec)
INFO:tensorflow:lr = 0.00011301667 (7.976 sec)
INFO:tensorflow:global_step/sec: 12.7577
INFO:tensorflow:loss = 1.2469722, step = 125860 (7.834 sec)
INFO:tensorflow:lr = 0.00011290314 (7.837 sec)
INFO:tensorflow:global_step/sec: 12.811
INFO:tensorflow:loss = 1.2492265, step = 125960 (7.806 sec)
INFO:tensorflow:lr = 0.00011278975 (7.802 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0305
INFO:tensorflow:loss = 1.2433882, step = 126060 (8.318 sec)
INFO:tensorflow:lr = 0.00011267645 (8.320 sec)
INFO:tensorflow:global_step/sec: 11.1724
INFO:tensorflow:loss = 1.2503134, step = 126160 (8.950 sec)
INFO:tensorflow:lr = 0.00011256325 (8.948 sec)
INFO:tensorflow:global_step/sec: 12.5939
INFO:tensorflow:loss = 1.2438878, step = 126260 (7.936 sec)
INFO:tensorflow:lr = 0.00011245018 (7.937 sec)
INFO:tensorflow:global_step/sec: 12.5034
INFO:tensorflow:loss = 1.2491252, step = 126360 (7.995 sec)
INFO:tensorflow:lr = 0.00011233722 (7.994 sec)
INFO:tensorflow:global_step/sec: 12.7607
INFO:tensorflow:loss = 1.2450532, step = 126460 (7.837 sec)
INFO:tensorflow:lr = 0.00011222437 (7.837 sec)
INFO:tensorflow:global_step/sec: 12.9278
INFO:tensorflow:loss = 1.2452111, step = 126560 (7.740 sec)
INFO:tensorflow:lr = 0.00011211164 (7.740 sec)
INFO:tensorflow:global_step/sec: 12.5386
INFO:tensorflow:loss = 1.2478788, step = 126660 (7.971 sec)
INFO:tensorflow:lr = 0.00011199902 (7.971 sec)
INFO:tensorflow:global_step/sec: 12.885
INFO:tensorflow:loss = 1.2447201, step = 126760 (7.761 sec)
INFO:tensorflow:lr = 0.00011188651 (7.766 sec)
INFO:tensorflow:global_step/sec: 12.4045
INFO:tensorflow:loss = 1.2440642, step = 126860 (8.065 sec)
INFO:tensorflow:lr = 0.000111774105 (8.060 sec)
INFO:tensorflow:global_step/sec: 12.7656
INFO:tensorflow:loss = 1.2417755, step = 126960 (7.830 sec)
INFO:tensorflow:lr = 0.00011166184 (7.830 sec)
INFO:tensorflow:global_step/sec: 12.4535
INFO:tensorflow:loss = 1.2505175, step = 127060 (8.031 sec)
INFO:tensorflow:lr = 0.00011154967 (8.031 sec)
INFO:tensorflow:Saving checkpoints for 127140 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2458371.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-127140
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-127140
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.724
INFO:tensorflow:Best (Exact Match) Accuracy: 0.724
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-127140
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 127140 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2435741, step = 127140
INFO:tensorflow:lr = 0.000111460016
INFO:tensorflow:global_step/sec: 10.2235
INFO:tensorflow:loss = 1.2465057, step = 127240 (9.783 sec)
INFO:tensorflow:lr = 0.000111348054 (9.788 sec)
INFO:tensorflow:global_step/sec: 12.5664
INFO:tensorflow:loss = 1.2451212, step = 127340 (7.958 sec)
INFO:tensorflow:lr = 0.0001112362 (7.953 sec)
INFO:tensorflow:global_step/sec: 12.5336
INFO:tensorflow:loss = 1.2496941, step = 127440 (7.979 sec)
INFO:tensorflow:lr = 0.000111124464 (7.979 sec)
INFO:tensorflow:global_step/sec: 12.8346
INFO:tensorflow:loss = 1.2417659, step = 127540 (7.791 sec)
INFO:tensorflow:lr = 0.000111012836 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.6779
INFO:tensorflow:loss = 1.2437657, step = 127640 (7.892 sec)
INFO:tensorflow:lr = 0.00011090133 (7.895 sec)
INFO:tensorflow:global_step/sec: 12.7347
INFO:tensorflow:loss = 1.2443695, step = 127740 (7.848 sec)
INFO:tensorflow:lr = 0.000110789915 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.4271
INFO:tensorflow:loss = 1.2493587, step = 127840 (8.049 sec)
INFO:tensorflow:lr = 0.00011067862 (8.051 sec)
INFO:tensorflow:global_step/sec: 12.5781
INFO:tensorflow:loss = 1.2441834, step = 127940 (7.953 sec)
INFO:tensorflow:lr = 0.00011056746 (7.951 sec)
INFO:tensorflow:global_step/sec: 12.8326
INFO:tensorflow:loss = 1.2549399, step = 128040 (7.792 sec)
INFO:tensorflow:lr = 0.00011045637 (7.792 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.24998
INFO:tensorflow:loss = 1.2466215, step = 128140 (12.121 sec)
INFO:tensorflow:lr = 0.00011034542 (12.123 sec)
INFO:tensorflow:global_step/sec: 12.5389
INFO:tensorflow:loss = 1.2452357, step = 128240 (7.971 sec)
INFO:tensorflow:lr = 0.00011023458 (7.974 sec)
INFO:tensorflow:global_step/sec: 12.4399
INFO:tensorflow:loss = 1.2436724, step = 128340 (8.043 sec)
INFO:tensorflow:lr = 0.00011012384 (8.039 sec)
INFO:tensorflow:global_step/sec: 12.2715
INFO:tensorflow:loss = 1.2544248, step = 128440 (8.145 sec)
INFO:tensorflow:lr = 0.00011001323 (8.145 sec)
INFO:tensorflow:global_step/sec: 12.527
INFO:tensorflow:loss = 1.2489731, step = 128540 (7.989 sec)
INFO:tensorflow:lr = 0.0001099027 (7.989 sec)
INFO:tensorflow:global_step/sec: 12.6973
INFO:tensorflow:loss = 1.2446793, step = 128640 (7.874 sec)
INFO:tensorflow:lr = 0.000109792316 (7.876 sec)
INFO:tensorflow:global_step/sec: 12.4341
INFO:tensorflow:loss = 1.2533972, step = 128740 (8.037 sec)
INFO:tensorflow:lr = 0.00010968202 (8.036 sec)
INFO:tensorflow:global_step/sec: 12.5414
INFO:tensorflow:loss = 1.2431896, step = 128840 (7.974 sec)
INFO:tensorflow:lr = 0.00010957183 (7.974 sec)
INFO:tensorflow:global_step/sec: 12.3855
INFO:tensorflow:loss = 1.2443397, step = 128940 (8.080 sec)
INFO:tensorflow:lr = 0.00010946177 (8.080 sec)
INFO:tensorflow:global_step/sec: 12.488
INFO:tensorflow:loss = 1.244244, step = 129040 (8.002 sec)
INFO:tensorflow:lr = 0.00010935181 (8.003 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.07857
INFO:tensorflow:loss = 1.2466154, step = 129140 (12.382 sec)
INFO:tensorflow:lr = 0.00010924197 (12.382 sec)
INFO:tensorflow:global_step/sec: 12.3629
INFO:tensorflow:loss = 1.2436008, step = 129240 (8.086 sec)
INFO:tensorflow:lr = 0.00010913224 (8.086 sec)
INFO:tensorflow:global_step/sec: 12.295
INFO:tensorflow:loss = 1.2524009, step = 129340 (8.135 sec)
INFO:tensorflow:lr = 0.00010902259 (8.135 sec)
INFO:tensorflow:global_step/sec: 12.3022
INFO:tensorflow:loss = 1.244007, step = 129440 (8.129 sec)
INFO:tensorflow:lr = 0.00010891308 (8.129 sec)
INFO:tensorflow:global_step/sec: 12.4037
INFO:tensorflow:loss = 1.2510656, step = 129540 (8.064 sec)
INFO:tensorflow:lr = 0.00010880369 (8.063 sec)
INFO:tensorflow:global_step/sec: 12.3284
INFO:tensorflow:loss = 1.2440971, step = 129640 (8.111 sec)
INFO:tensorflow:lr = 0.000108694396 (8.111 sec)
INFO:tensorflow:global_step/sec: 12.1042
INFO:tensorflow:loss = 1.2430845, step = 129740 (8.262 sec)
INFO:tensorflow:lr = 0.00010858521 (8.262 sec)
INFO:tensorflow:global_step/sec: 12.1477
INFO:tensorflow:loss = 1.2497059, step = 129840 (8.226 sec)
INFO:tensorflow:lr = 0.000108476124 (8.227 sec)
INFO:tensorflow:global_step/sec: 12.1074
INFO:tensorflow:loss = 1.2511698, step = 129940 (8.260 sec)
INFO:tensorflow:lr = 0.00010836717 (8.258 sec)
INFO:tensorflow:global_step/sec: 12.6255
INFO:tensorflow:loss = 1.2424986, step = 130040 (7.926 sec)
INFO:tensorflow:lr = 0.000108258304 (7.927 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.91354
INFO:tensorflow:loss = 1.2430753, step = 130140 (12.632 sec)
INFO:tensorflow:lr = 0.000108149565 (12.635 sec)
INFO:tensorflow:global_step/sec: 12.3666
INFO:tensorflow:loss = 1.2508368, step = 130240 (8.087 sec)
INFO:tensorflow:lr = 0.00010804091 (8.087 sec)
INFO:tensorflow:global_step/sec: 12.3495
INFO:tensorflow:loss = 1.2460352, step = 130340 (8.097 sec)
INFO:tensorflow:lr = 0.00010793238 (8.092 sec)
INFO:tensorflow:global_step/sec: 12.2387
INFO:tensorflow:loss = 1.2445351, step = 130440 (8.175 sec)
INFO:tensorflow:lr = 0.000107823966 (8.175 sec)
INFO:tensorflow:global_step/sec: 12.2714
INFO:tensorflow:loss = 1.24298, step = 130540 (8.144 sec)
INFO:tensorflow:lr = 0.00010771565 (8.149 sec)
INFO:tensorflow:global_step/sec: 12.2735
INFO:tensorflow:loss = 1.2450466, step = 130640 (8.149 sec)
INFO:tensorflow:lr = 0.000107607455 (8.145 sec)
INFO:tensorflow:global_step/sec: 12.2928
INFO:tensorflow:loss = 1.2488058, step = 130740 (8.135 sec)
INFO:tensorflow:lr = 0.00010749935 (8.136 sec)
INFO:tensorflow:global_step/sec: 12.1511
INFO:tensorflow:loss = 1.2458876, step = 130840 (8.228 sec)
INFO:tensorflow:lr = 0.00010739135 (8.227 sec)
INFO:tensorflow:global_step/sec: 12.4498
INFO:tensorflow:loss = 1.2464055, step = 130940 (8.038 sec)
INFO:tensorflow:lr = 0.00010728349 (8.038 sec)
INFO:tensorflow:global_step/sec: 12.5961
INFO:tensorflow:loss = 1.2468756, step = 131040 (7.933 sec)
INFO:tensorflow:lr = 0.000107175714 (7.934 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.15478
INFO:tensorflow:loss = 1.2451632, step = 131140 (12.266 sec)
INFO:tensorflow:lr = 0.000107068074 (12.265 sec)
INFO:tensorflow:global_step/sec: 12.3205
INFO:tensorflow:loss = 1.2453297, step = 131240 (8.116 sec)
INFO:tensorflow:lr = 0.0001069605 (8.117 sec)
INFO:tensorflow:global_step/sec: 11.9643
INFO:tensorflow:loss = 1.2488296, step = 131340 (8.361 sec)
INFO:tensorflow:lr = 0.000106853055 (8.360 sec)
INFO:tensorflow:global_step/sec: 12.4314
INFO:tensorflow:loss = 1.2435274, step = 131440 (8.043 sec)
INFO:tensorflow:lr = 0.00010674573 (8.043 sec)
INFO:tensorflow:global_step/sec: 12.0054
INFO:tensorflow:loss = 1.245277, step = 131540 (8.327 sec)
INFO:tensorflow:lr = 0.0001066385 (8.328 sec)
INFO:tensorflow:global_step/sec: 12.2444
INFO:tensorflow:loss = 1.243785, step = 131640 (8.168 sec)
INFO:tensorflow:lr = 0.00010653138 (8.172 sec)
INFO:tensorflow:global_step/sec: 12.2932
INFO:tensorflow:loss = 1.248311, step = 131740 (8.131 sec)
INFO:tensorflow:lr = 0.00010642436 (8.126 sec)
INFO:tensorflow:global_step/sec: 12.3766
INFO:tensorflow:loss = 1.2436168, step = 131840 (8.086 sec)
INFO:tensorflow:lr = 0.00010631745 (8.086 sec)
INFO:tensorflow:global_step/sec: 12.3664
INFO:tensorflow:loss = 1.2433931, step = 131940 (8.088 sec)
INFO:tensorflow:lr = 0.00010621067 (8.088 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.14088
INFO:tensorflow:loss = 1.246407, step = 132040 (12.280 sec)
INFO:tensorflow:lr = 0.00010610395 (12.281 sec)
INFO:tensorflow:global_step/sec: 12.2325
INFO:tensorflow:loss = 1.2487501, step = 132140 (8.171 sec)
INFO:tensorflow:lr = 0.00010599738 (8.171 sec)
INFO:tensorflow:global_step/sec: 12.167
INFO:tensorflow:loss = 1.2465798, step = 132240 (8.220 sec)
INFO:tensorflow:lr = 0.000105890904 (8.219 sec)
INFO:tensorflow:global_step/sec: 12.1129
INFO:tensorflow:loss = 1.2428634, step = 132340 (8.257 sec)
INFO:tensorflow:lr = 0.00010578453 (8.258 sec)
INFO:tensorflow:global_step/sec: 12.1236
INFO:tensorflow:loss = 1.248153, step = 132440 (8.246 sec)
INFO:tensorflow:lr = 0.00010567827 (8.250 sec)
INFO:tensorflow:global_step/sec: 12.2855
INFO:tensorflow:loss = 1.2455584, step = 132540 (8.139 sec)
INFO:tensorflow:lr = 0.0001055721 (8.135 sec)
INFO:tensorflow:global_step/sec: 12.1064
INFO:tensorflow:loss = 1.244128, step = 132640 (8.263 sec)
INFO:tensorflow:lr = 0.00010546607 (8.263 sec)
INFO:tensorflow:global_step/sec: 12.0689
INFO:tensorflow:loss = 1.2437512, step = 132740 (8.283 sec)
INFO:tensorflow:lr = 0.00010536012 (8.283 sec)
INFO:tensorflow:global_step/sec: 12.582
INFO:tensorflow:loss = 1.2527406, step = 132840 (7.953 sec)
INFO:tensorflow:lr = 0.00010525428 (7.960 sec)
INFO:tensorflow:global_step/sec: 12.4056
INFO:tensorflow:loss = 1.2459419, step = 132940 (8.061 sec)
INFO:tensorflow:lr = 0.00010514854 (8.055 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.60371
INFO:tensorflow:loss = 1.2440192, step = 133040 (11.618 sec)
INFO:tensorflow:lr = 0.00010504292 (11.617 sec)
INFO:tensorflow:global_step/sec: 12.3615
INFO:tensorflow:loss = 1.246831, step = 133140 (8.095 sec)
INFO:tensorflow:lr = 0.000104937404 (8.096 sec)
INFO:tensorflow:global_step/sec: 12.1262
INFO:tensorflow:loss = 1.2434736, step = 133240 (8.243 sec)
INFO:tensorflow:lr = 0.00010483199 (8.244 sec)
INFO:tensorflow:global_step/sec: 11.9553
INFO:tensorflow:loss = 1.2503418, step = 133340 (8.365 sec)
INFO:tensorflow:lr = 0.00010472667 (8.364 sec)
INFO:tensorflow:global_step/sec: 11.8907
INFO:tensorflow:loss = 1.2424436, step = 133440 (8.409 sec)
INFO:tensorflow:lr = 0.00010462148 (8.408 sec)
INFO:tensorflow:global_step/sec: 12.2531
INFO:tensorflow:loss = 1.2495891, step = 133540 (8.160 sec)
INFO:tensorflow:lr = 0.000104516395 (8.160 sec)
INFO:tensorflow:global_step/sec: 12.2438
INFO:tensorflow:loss = 1.2514443, step = 133640 (8.172 sec)
INFO:tensorflow:lr = 0.00010441141 (8.174 sec)
INFO:tensorflow:global_step/sec: 12.1721
INFO:tensorflow:loss = 1.2426683, step = 133740 (8.216 sec)
INFO:tensorflow:lr = 0.00010430651 (8.216 sec)
INFO:tensorflow:global_step/sec: 12.3861
INFO:tensorflow:loss = 1.2496032, step = 133840 (8.072 sec)
INFO:tensorflow:lr = 0.00010420173 (8.072 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.7193
INFO:tensorflow:loss = 1.2486618, step = 133940 (8.536 sec)
INFO:tensorflow:lr = 0.00010409707 (8.537 sec)
INFO:tensorflow:global_step/sec: 10.1707
INFO:tensorflow:loss = 1.2442518, step = 134040 (9.830 sec)
INFO:tensorflow:lr = 0.0001039925 (9.829 sec)
INFO:tensorflow:global_step/sec: 12.3223
INFO:tensorflow:loss = 1.2536051, step = 134140 (8.114 sec)
INFO:tensorflow:lr = 0.00010388803 (8.115 sec)
INFO:tensorflow:global_step/sec: 12.4179
INFO:tensorflow:loss = 1.2437172, step = 134240 (8.052 sec)
INFO:tensorflow:lr = 0.00010378366 (8.053 sec)
INFO:tensorflow:global_step/sec: 12.2421
INFO:tensorflow:loss = 1.243759, step = 134340 (8.171 sec)
INFO:tensorflow:lr = 0.000103679406 (8.169 sec)
INFO:tensorflow:global_step/sec: 12.1969
INFO:tensorflow:loss = 1.2443966, step = 134440 (8.194 sec)
INFO:tensorflow:lr = 0.000103575265 (8.195 sec)
INFO:tensorflow:global_step/sec: 12.5337
INFO:tensorflow:loss = 1.2484913, step = 134540 (7.979 sec)
INFO:tensorflow:lr = 0.000103471226 (7.979 sec)
INFO:tensorflow:global_step/sec: 12.4363
INFO:tensorflow:loss = 1.2449461, step = 134640 (8.041 sec)
INFO:tensorflow:lr = 0.00010336729 (8.041 sec)
INFO:tensorflow:global_step/sec: 12.1351
INFO:tensorflow:loss = 1.2493279, step = 134740 (8.239 sec)
INFO:tensorflow:lr = 0.00010326345 (8.239 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.9634
INFO:tensorflow:loss = 1.246514, step = 134840 (8.360 sec)
INFO:tensorflow:lr = 0.00010315971 (8.363 sec)
INFO:tensorflow:global_step/sec: 10.6141
INFO:tensorflow:loss = 1.2479151, step = 134940 (9.421 sec)
INFO:tensorflow:lr = 0.000103056096 (9.418 sec)
INFO:tensorflow:global_step/sec: 12.1932
INFO:tensorflow:loss = 1.2446302, step = 135040 (8.205 sec)
INFO:tensorflow:lr = 0.000102952574 (8.206 sec)
INFO:tensorflow:global_step/sec: 12.3143
INFO:tensorflow:loss = 1.2450219, step = 135140 (8.120 sec)
INFO:tensorflow:lr = 0.00010284917 (8.119 sec)
INFO:tensorflow:global_step/sec: 12.2555
INFO:tensorflow:loss = 1.2433691, step = 135240 (8.155 sec)
INFO:tensorflow:lr = 0.00010274584 (8.155 sec)
INFO:tensorflow:global_step/sec: 12.2253
INFO:tensorflow:loss = 1.2495849, step = 135340 (8.180 sec)
INFO:tensorflow:lr = 0.00010264262 (8.180 sec)
INFO:tensorflow:global_step/sec: 12.225
INFO:tensorflow:loss = 1.2457464, step = 135440 (8.185 sec)
INFO:tensorflow:lr = 0.000102539525 (8.185 sec)
INFO:tensorflow:global_step/sec: 12.3318
INFO:tensorflow:loss = 1.2438371, step = 135540 (8.107 sec)
INFO:tensorflow:lr = 0.00010243652 (8.107 sec)
INFO:tensorflow:global_step/sec: 12.4329
INFO:tensorflow:loss = 1.2439129, step = 135640 (8.047 sec)
INFO:tensorflow:lr = 0.000102333615 (8.047 sec)
INFO:tensorflow:global_step/sec: 12.4265
INFO:tensorflow:loss = 1.2445234, step = 135740 (8.041 sec)
INFO:tensorflow:lr = 0.00010223082 (8.042 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.3109
INFO:tensorflow:loss = 1.247158, step = 135840 (8.856 sec)
INFO:tensorflow:lr = 0.00010212812 (8.856 sec)
INFO:tensorflow:global_step/sec: 10.9069
INFO:tensorflow:loss = 1.2456703, step = 135940 (9.154 sec)
INFO:tensorflow:lr = 0.000102025544 (9.158 sec)
INFO:tensorflow:global_step/sec: 12.004
INFO:tensorflow:loss = 1.249978, step = 136040 (8.336 sec)
INFO:tensorflow:lr = 0.00010192303 (8.331 sec)
INFO:tensorflow:global_step/sec: 12.3161
INFO:tensorflow:loss = 1.2440797, step = 136140 (8.115 sec)
INFO:tensorflow:lr = 0.00010182066 (8.115 sec)
INFO:tensorflow:global_step/sec: 12.4287
INFO:tensorflow:loss = 1.2494134, step = 136240 (8.046 sec)
INFO:tensorflow:lr = 0.000101718375 (8.046 sec)
INFO:tensorflow:global_step/sec: 12.2461
INFO:tensorflow:loss = 1.2424796, step = 136340 (8.169 sec)
INFO:tensorflow:lr = 0.0001016162 (8.171 sec)
INFO:tensorflow:global_step/sec: 12.1615
INFO:tensorflow:loss = 1.2436228, step = 136440 (8.224 sec)
INFO:tensorflow:lr = 0.00010151413 (8.221 sec)
INFO:tensorflow:global_step/sec: 12.4358
INFO:tensorflow:loss = 1.2450893, step = 136540 (8.037 sec)
INFO:tensorflow:lr = 0.00010141214 (8.038 sec)
INFO:tensorflow:global_step/sec: 12.3499
INFO:tensorflow:loss = 1.2470615, step = 136640 (8.102 sec)
INFO:tensorflow:lr = 0.000101310274 (8.101 sec)
INFO:tensorflow:global_step/sec: 12.0695
INFO:tensorflow:loss = 1.2428521, step = 136740 (8.283 sec)
INFO:tensorflow:lr = 0.000101208505 (8.284 sec)
INFO:tensorflow:global_step/sec: 12.4794
INFO:tensorflow:loss = 1.243703, step = 136840 (8.011 sec)
INFO:tensorflow:lr = 0.000101106845 (8.011 sec)
INFO:tensorflow:Saving checkpoints for 136920 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2554393.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-136920
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-136920
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.723
INFO:tensorflow:Best (Exact Match) Accuracy: 0.724
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-136920
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 136920 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2442503, step = 136920
INFO:tensorflow:lr = 0.00010102558
INFO:tensorflow:global_step/sec: 9.49728
INFO:tensorflow:loss = 1.2426513, step = 137020 (10.534 sec)
INFO:tensorflow:lr = 0.00010092411 (10.534 sec)
INFO:tensorflow:global_step/sec: 12.1446
INFO:tensorflow:loss = 1.2465923, step = 137120 (8.235 sec)
INFO:tensorflow:lr = 0.00010082272 (8.235 sec)
INFO:tensorflow:global_step/sec: 12.0058
INFO:tensorflow:loss = 1.2437319, step = 137220 (8.331 sec)
INFO:tensorflow:lr = 0.00010072145 (8.333 sec)
INFO:tensorflow:global_step/sec: 12.1429
INFO:tensorflow:loss = 1.2430804, step = 137320 (8.237 sec)
INFO:tensorflow:lr = 0.000100620266 (8.234 sec)
INFO:tensorflow:global_step/sec: 11.9963
INFO:tensorflow:loss = 1.2503809, step = 137420 (8.329 sec)
INFO:tensorflow:lr = 0.00010051919 (8.330 sec)
INFO:tensorflow:global_step/sec: 11.9954
INFO:tensorflow:loss = 1.245267, step = 137520 (8.336 sec)
INFO:tensorflow:lr = 0.00010041822 (8.345 sec)
INFO:tensorflow:global_step/sec: 12.0687
INFO:tensorflow:loss = 1.244724, step = 137620 (8.290 sec)
INFO:tensorflow:lr = 0.00010031733 (8.283 sec)
INFO:tensorflow:global_step/sec: 12.3354
INFO:tensorflow:loss = 1.2446139, step = 137720 (8.108 sec)
INFO:tensorflow:lr = 0.000100216566 (8.107 sec)
INFO:tensorflow:global_step/sec: 12.2787
INFO:tensorflow:loss = 1.2453194, step = 137820 (8.139 sec)
INFO:tensorflow:lr = 0.0001001159 (8.138 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.83209
INFO:tensorflow:loss = 1.2472328, step = 137920 (12.767 sec)
INFO:tensorflow:lr = 0.00010001531 (12.769 sec)
INFO:tensorflow:global_step/sec: 12.1816
INFO:tensorflow:loss = 1.2472757, step = 138020 (8.214 sec)
INFO:tensorflow:lr = 9.991486e-05 (8.212 sec)
INFO:tensorflow:global_step/sec: 12.2596
INFO:tensorflow:loss = 1.2465109, step = 138120 (8.152 sec)
INFO:tensorflow:lr = 9.981449e-05 (8.158 sec)
INFO:tensorflow:global_step/sec: 12.3647
INFO:tensorflow:loss = 1.2436755, step = 138220 (8.094 sec)
INFO:tensorflow:lr = 9.971423e-05 (8.088 sec)
INFO:tensorflow:global_step/sec: 12.373
INFO:tensorflow:loss = 1.2455561, step = 138320 (8.083 sec)
INFO:tensorflow:lr = 9.961406e-05 (8.086 sec)
INFO:tensorflow:global_step/sec: 12.1114
INFO:tensorflow:loss = 1.2426962, step = 138420 (8.256 sec)
INFO:tensorflow:lr = 9.9513985e-05 (8.254 sec)
INFO:tensorflow:global_step/sec: 12.3349
INFO:tensorflow:loss = 1.2463111, step = 138520 (8.101 sec)
INFO:tensorflow:lr = 9.9414036e-05 (8.105 sec)
INFO:tensorflow:global_step/sec: 12.0076
INFO:tensorflow:loss = 1.2430253, step = 138620 (8.328 sec)
INFO:tensorflow:lr = 9.9314166e-05 (8.329 sec)
INFO:tensorflow:global_step/sec: 12.1418
INFO:tensorflow:loss = 1.2426336, step = 138720 (8.243 sec)
INFO:tensorflow:lr = 9.9214405e-05 (8.238 sec)
INFO:tensorflow:global_step/sec: 12.0151
INFO:tensorflow:loss = 1.2564312, step = 138820 (8.317 sec)
INFO:tensorflow:lr = 9.911473e-05 (8.316 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.72
INFO:tensorflow:loss = 1.2428147, step = 138920 (12.956 sec)
INFO:tensorflow:lr = 9.901517e-05 (12.958 sec)
INFO:tensorflow:global_step/sec: 12.1405
INFO:tensorflow:loss = 1.2480109, step = 139020 (8.235 sec)
INFO:tensorflow:lr = 9.891571e-05 (8.235 sec)
INFO:tensorflow:global_step/sec: 12.1696
INFO:tensorflow:loss = 1.2434272, step = 139120 (8.221 sec)
INFO:tensorflow:lr = 9.8816345e-05 (8.220 sec)
INFO:tensorflow:global_step/sec: 12.3821
INFO:tensorflow:loss = 1.2433711, step = 139220 (8.073 sec)
INFO:tensorflow:lr = 9.871709e-05 (8.073 sec)
INFO:tensorflow:global_step/sec: 11.899
INFO:tensorflow:loss = 1.2432872, step = 139320 (8.402 sec)
INFO:tensorflow:lr = 9.861792e-05 (8.401 sec)
INFO:tensorflow:global_step/sec: 12.294
INFO:tensorflow:loss = 1.2443063, step = 139420 (8.140 sec)
INFO:tensorflow:lr = 9.8518845e-05 (8.140 sec)
INFO:tensorflow:global_step/sec: 11.9733
INFO:tensorflow:loss = 1.2486244, step = 139520 (8.351 sec)
INFO:tensorflow:lr = 9.841989e-05 (8.351 sec)
INFO:tensorflow:global_step/sec: 12.3805
INFO:tensorflow:loss = 1.2429479, step = 139620 (8.074 sec)
INFO:tensorflow:lr = 9.832102e-05 (8.075 sec)
INFO:tensorflow:global_step/sec: 12.2166
INFO:tensorflow:loss = 1.2444541, step = 139720 (8.189 sec)
INFO:tensorflow:lr = 9.822226e-05 (8.189 sec)
INFO:tensorflow:global_step/sec: 12.0897
INFO:tensorflow:loss = 1.2436047, step = 139820 (8.271 sec)
INFO:tensorflow:lr = 9.8123586e-05 (8.271 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.94567
INFO:tensorflow:loss = 1.2447098, step = 139920 (12.585 sec)
INFO:tensorflow:lr = 9.802502e-05 (12.584 sec)
INFO:tensorflow:global_step/sec: 12.1545
INFO:tensorflow:loss = 1.2488983, step = 140020 (8.230 sec)
INFO:tensorflow:lr = 9.792656e-05 (8.232 sec)
INFO:tensorflow:global_step/sec: 12.3331
INFO:tensorflow:loss = 1.2468932, step = 140120 (8.102 sec)
INFO:tensorflow:lr = 9.782818e-05 (8.103 sec)
INFO:tensorflow:global_step/sec: 12.1976
INFO:tensorflow:loss = 1.2424612, step = 140220 (8.202 sec)
INFO:tensorflow:lr = 9.772992e-05 (8.198 sec)
INFO:tensorflow:global_step/sec: 12.1431
INFO:tensorflow:loss = 1.245854, step = 140320 (8.232 sec)
INFO:tensorflow:lr = 9.763174e-05 (8.232 sec)
INFO:tensorflow:global_step/sec: 12.2553
INFO:tensorflow:loss = 1.2430899, step = 140420 (8.160 sec)
INFO:tensorflow:lr = 9.753366e-05 (8.162 sec)
INFO:tensorflow:global_step/sec: 12.3075
INFO:tensorflow:loss = 1.2432994, step = 140520 (8.129 sec)
INFO:tensorflow:lr = 9.7435695e-05 (8.127 sec)
INFO:tensorflow:global_step/sec: 12.2089
INFO:tensorflow:loss = 1.2422528, step = 140620 (8.190 sec)
INFO:tensorflow:lr = 9.733782e-05 (8.189 sec)
INFO:tensorflow:global_step/sec: 12.4876
INFO:tensorflow:loss = 1.2459801, step = 140720 (8.004 sec)
INFO:tensorflow:lr = 9.7240045e-05 (8.005 sec)
INFO:tensorflow:global_step/sec: 12.246
INFO:tensorflow:loss = 1.2448704, step = 140820 (8.166 sec)
INFO:tensorflow:lr = 9.714235e-05 (8.166 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.94083
INFO:tensorflow:loss = 1.2432497, step = 140920 (12.593 sec)
INFO:tensorflow:lr = 9.7044765e-05 (12.594 sec)
INFO:tensorflow:global_step/sec: 12.578
INFO:tensorflow:loss = 1.2454319, step = 141020 (7.955 sec)
INFO:tensorflow:lr = 9.694729e-05 (7.955 sec)
INFO:tensorflow:global_step/sec: 12.2807
INFO:tensorflow:loss = 1.2427717, step = 141120 (8.142 sec)
INFO:tensorflow:lr = 9.68499e-05 (8.141 sec)
INFO:tensorflow:global_step/sec: 12.3299
INFO:tensorflow:loss = 1.2423016, step = 141220 (8.115 sec)
INFO:tensorflow:lr = 9.6752614e-05 (8.119 sec)
INFO:tensorflow:global_step/sec: 12.2827
INFO:tensorflow:loss = 1.2446163, step = 141320 (8.139 sec)
INFO:tensorflow:lr = 9.665543e-05 (8.135 sec)
INFO:tensorflow:global_step/sec: 12.4412
INFO:tensorflow:loss = 1.2545216, step = 141420 (8.038 sec)
INFO:tensorflow:lr = 9.655833e-05 (8.038 sec)
INFO:tensorflow:global_step/sec: 12.4062
INFO:tensorflow:loss = 1.2451398, step = 141520 (8.059 sec)
INFO:tensorflow:lr = 9.646134e-05 (8.059 sec)
INFO:tensorflow:global_step/sec: 12.2396
INFO:tensorflow:loss = 1.2458344, step = 141620 (8.166 sec)
INFO:tensorflow:lr = 9.6364434e-05 (8.171 sec)
INFO:tensorflow:global_step/sec: 12.5565
INFO:tensorflow:loss = 1.2435321, step = 141720 (7.964 sec)
INFO:tensorflow:lr = 9.626764e-05 (7.959 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.42035
INFO:tensorflow:loss = 1.2429127, step = 141820 (11.878 sec)
INFO:tensorflow:lr = 9.6170945e-05 (11.878 sec)
INFO:tensorflow:global_step/sec: 12.3013
INFO:tensorflow:loss = 1.2506694, step = 141920 (8.127 sec)
INFO:tensorflow:lr = 9.607432e-05 (8.126 sec)
INFO:tensorflow:global_step/sec: 12.7147
INFO:tensorflow:loss = 1.2428731, step = 142020 (7.865 sec)
INFO:tensorflow:lr = 9.597782e-05 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.5149
INFO:tensorflow:loss = 1.2532231, step = 142120 (7.991 sec)
INFO:tensorflow:lr = 9.5881405e-05 (7.990 sec)
INFO:tensorflow:global_step/sec: 12.3754
INFO:tensorflow:loss = 1.2438812, step = 142220 (8.080 sec)
INFO:tensorflow:lr = 9.578509e-05 (8.081 sec)
INFO:tensorflow:global_step/sec: 12.5211
INFO:tensorflow:loss = 1.2414047, step = 142320 (7.986 sec)
INFO:tensorflow:lr = 9.568886e-05 (7.985 sec)
INFO:tensorflow:global_step/sec: 12.4172
INFO:tensorflow:loss = 1.2448086, step = 142420 (8.059 sec)
INFO:tensorflow:lr = 9.559275e-05 (8.059 sec)
INFO:tensorflow:global_step/sec: 12.5996
INFO:tensorflow:loss = 1.241774, step = 142520 (7.939 sec)
INFO:tensorflow:lr = 9.549673e-05 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.1413
INFO:tensorflow:loss = 1.2439814, step = 142620 (8.235 sec)
INFO:tensorflow:lr = 9.540079e-05 (8.232 sec)
INFO:tensorflow:global_step/sec: 12.2161
INFO:tensorflow:loss = 1.2453963, step = 142720 (8.186 sec)
INFO:tensorflow:lr = 9.530496e-05 (8.185 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.8836
INFO:tensorflow:loss = 1.2426373, step = 142820 (11.257 sec)
INFO:tensorflow:lr = 9.520922e-05 (11.256 sec)
INFO:tensorflow:global_step/sec: 12.6974
INFO:tensorflow:loss = 1.2442544, step = 142920 (7.873 sec)
INFO:tensorflow:lr = 9.511357e-05 (7.875 sec)
INFO:tensorflow:global_step/sec: 12.4545
INFO:tensorflow:loss = 1.2444447, step = 143020 (8.030 sec)
INFO:tensorflow:lr = 9.501804e-05 (8.029 sec)
INFO:tensorflow:global_step/sec: 12.2781
INFO:tensorflow:loss = 1.2456886, step = 143120 (8.145 sec)
INFO:tensorflow:lr = 9.492259e-05 (8.146 sec)
INFO:tensorflow:global_step/sec: 12.6858
INFO:tensorflow:loss = 1.2470523, step = 143220 (7.878 sec)
INFO:tensorflow:lr = 9.482725e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.4909
INFO:tensorflow:loss = 1.2446479, step = 143320 (8.006 sec)
INFO:tensorflow:lr = 9.4731986e-05 (8.010 sec)
INFO:tensorflow:global_step/sec: 12.3774
INFO:tensorflow:loss = 1.2446898, step = 143420 (8.085 sec)
INFO:tensorflow:lr = 9.4636816e-05 (8.080 sec)
INFO:tensorflow:global_step/sec: 12.4325
INFO:tensorflow:loss = 1.2465398, step = 143520 (8.043 sec)
INFO:tensorflow:lr = 9.454176e-05 (8.043 sec)
INFO:tensorflow:global_step/sec: 12.2342
INFO:tensorflow:loss = 1.2445402, step = 143620 (8.173 sec)
INFO:tensorflow:lr = 9.444679e-05 (8.173 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1484
INFO:tensorflow:loss = 1.2433293, step = 143720 (8.235 sec)
INFO:tensorflow:lr = 9.435192e-05 (8.237 sec)
INFO:tensorflow:global_step/sec: 10.611
INFO:tensorflow:loss = 1.2452755, step = 143820 (9.419 sec)
INFO:tensorflow:lr = 9.4257135e-05 (9.417 sec)
INFO:tensorflow:global_step/sec: 12.7544
INFO:tensorflow:loss = 1.2486116, step = 143920 (7.843 sec)
INFO:tensorflow:lr = 9.4162446e-05 (7.843 sec)
INFO:tensorflow:global_step/sec: 12.6621
INFO:tensorflow:loss = 1.2436063, step = 144020 (7.892 sec)
INFO:tensorflow:lr = 9.4067866e-05 (7.892 sec)
INFO:tensorflow:global_step/sec: 12.3106
INFO:tensorflow:loss = 1.2431039, step = 144120 (8.124 sec)
INFO:tensorflow:lr = 9.3973365e-05 (8.124 sec)
INFO:tensorflow:global_step/sec: 12.5707
INFO:tensorflow:loss = 1.2427852, step = 144220 (7.954 sec)
INFO:tensorflow:lr = 9.387897e-05 (7.953 sec)
INFO:tensorflow:global_step/sec: 12.4048
INFO:tensorflow:loss = 1.2442795, step = 144320 (8.063 sec)
INFO:tensorflow:lr = 9.3784656e-05 (8.064 sec)
INFO:tensorflow:global_step/sec: 12.5874
INFO:tensorflow:loss = 1.24356, step = 144420 (7.947 sec)
INFO:tensorflow:lr = 9.369045e-05 (7.947 sec)
INFO:tensorflow:global_step/sec: 12.514
INFO:tensorflow:loss = 1.2426248, step = 144520 (7.989 sec)
INFO:tensorflow:lr = 9.359634e-05 (7.988 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2544
INFO:tensorflow:loss = 1.2479446, step = 144620 (8.165 sec)
INFO:tensorflow:lr = 9.350232e-05 (8.166 sec)
INFO:tensorflow:global_step/sec: 10.9916
INFO:tensorflow:loss = 1.2435095, step = 144720 (9.092 sec)
INFO:tensorflow:lr = 9.3408395e-05 (9.091 sec)
INFO:tensorflow:global_step/sec: 12.7682
INFO:tensorflow:loss = 1.2506919, step = 144820 (7.832 sec)
INFO:tensorflow:lr = 9.3314564e-05 (7.831 sec)
INFO:tensorflow:global_step/sec: 12.4579
INFO:tensorflow:loss = 1.2430148, step = 144920 (8.028 sec)
INFO:tensorflow:lr = 9.322082e-05 (8.032 sec)
INFO:tensorflow:global_step/sec: 12.5404
INFO:tensorflow:loss = 1.2427722, step = 145020 (7.975 sec)
INFO:tensorflow:lr = 9.3127186e-05 (7.969 sec)
INFO:tensorflow:global_step/sec: 12.6
INFO:tensorflow:loss = 1.2560487, step = 145120 (7.941 sec)
INFO:tensorflow:lr = 9.303363e-05 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.4571
INFO:tensorflow:loss = 1.2459077, step = 145220 (8.022 sec)
INFO:tensorflow:lr = 9.294019e-05 (8.021 sec)
INFO:tensorflow:global_step/sec: 12.7962
INFO:tensorflow:loss = 1.2429833, step = 145320 (7.815 sec)
INFO:tensorflow:lr = 9.284682e-05 (7.816 sec)
INFO:tensorflow:global_step/sec: 12.6749
INFO:tensorflow:loss = 1.244671, step = 145420 (7.890 sec)
INFO:tensorflow:lr = 9.275354e-05 (7.893 sec)
INFO:tensorflow:global_step/sec: 12.7
INFO:tensorflow:loss = 1.2438098, step = 145520 (7.877 sec)
INFO:tensorflow:lr = 9.2660375e-05 (7.873 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.5416
INFO:tensorflow:loss = 1.2416648, step = 145620 (8.670 sec)
INFO:tensorflow:lr = 9.25673e-05 (8.670 sec)
INFO:tensorflow:global_step/sec: 11.4268
INFO:tensorflow:loss = 1.2421402, step = 145720 (8.746 sec)
INFO:tensorflow:lr = 9.2474314e-05 (8.750 sec)
INFO:tensorflow:global_step/sec: 12.6866
INFO:tensorflow:loss = 1.2447598, step = 145820 (7.886 sec)
INFO:tensorflow:lr = 9.2381415e-05 (7.881 sec)
INFO:tensorflow:global_step/sec: 12.6577
INFO:tensorflow:loss = 1.2479534, step = 145920 (7.901 sec)
INFO:tensorflow:lr = 9.228862e-05 (7.901 sec)
INFO:tensorflow:global_step/sec: 12.1214
INFO:tensorflow:loss = 1.2518696, step = 146020 (8.243 sec)
INFO:tensorflow:lr = 9.219592e-05 (8.245 sec)
INFO:tensorflow:global_step/sec: 12.8833
INFO:tensorflow:loss = 1.248869, step = 146120 (7.764 sec)
INFO:tensorflow:lr = 9.210329e-05 (7.762 sec)
INFO:tensorflow:global_step/sec: 12.4983
INFO:tensorflow:loss = 1.2434883, step = 146220 (7.999 sec)
INFO:tensorflow:lr = 9.2010785e-05 (8.001 sec)
INFO:tensorflow:global_step/sec: 12.7547
INFO:tensorflow:loss = 1.2447973, step = 146320 (7.843 sec)
INFO:tensorflow:lr = 9.1918344e-05 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.4445
INFO:tensorflow:loss = 1.246257, step = 146420 (8.035 sec)
INFO:tensorflow:lr = 9.182601e-05 (8.034 sec)
INFO:tensorflow:global_step/sec: 12.6804
INFO:tensorflow:loss = 1.2494022, step = 146520 (7.884 sec)
INFO:tensorflow:lr = 9.173378e-05 (7.883 sec)
INFO:tensorflow:global_step/sec: 12.6978
INFO:tensorflow:loss = 1.2443601, step = 146620 (7.877 sec)
INFO:tensorflow:lr = 9.164162e-05 (7.877 sec)
INFO:tensorflow:Saving checkpoints for 146700 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2431201.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-146700
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-146700
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.724
INFO:tensorflow:Best (Exact Match) Accuracy: 0.724
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-146700
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 146700 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2443894, step = 146700
INFO:tensorflow:lr = 9.156797e-05
INFO:tensorflow:global_step/sec: 10.051
INFO:tensorflow:loss = 1.244071, step = 146800 (9.952 sec)
INFO:tensorflow:lr = 9.1475995e-05 (9.956 sec)
INFO:tensorflow:global_step/sec: 12.6786
INFO:tensorflow:loss = 1.255505, step = 146900 (7.891 sec)
INFO:tensorflow:lr = 9.138411e-05 (7.887 sec)
INFO:tensorflow:global_step/sec: 12.6207
INFO:tensorflow:loss = 1.2456456, step = 147000 (7.924 sec)
INFO:tensorflow:lr = 9.129231e-05 (7.924 sec)
INFO:tensorflow:global_step/sec: 12.7906
INFO:tensorflow:loss = 1.2438047, step = 147100 (7.815 sec)
INFO:tensorflow:lr = 9.12006e-05 (7.816 sec)
INFO:tensorflow:global_step/sec: 12.5422
INFO:tensorflow:loss = 1.2472233, step = 147200 (7.972 sec)
INFO:tensorflow:lr = 9.1108974e-05 (7.971 sec)
INFO:tensorflow:global_step/sec: 12.6121
INFO:tensorflow:loss = 1.2505473, step = 147300 (7.928 sec)
INFO:tensorflow:lr = 9.101747e-05 (7.928 sec)
INFO:tensorflow:global_step/sec: 12.5697
INFO:tensorflow:loss = 1.245287, step = 147400 (7.956 sec)
INFO:tensorflow:lr = 9.092603e-05 (7.963 sec)
INFO:tensorflow:global_step/sec: 12.6833
INFO:tensorflow:loss = 1.2446585, step = 147500 (7.886 sec)
INFO:tensorflow:lr = 9.08347e-05 (7.880 sec)
INFO:tensorflow:global_step/sec: 12.8006
INFO:tensorflow:loss = 1.2419182, step = 147600 (7.811 sec)
INFO:tensorflow:lr = 9.0743444e-05 (7.811 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.31527
INFO:tensorflow:loss = 1.2440436, step = 147700 (12.030 sec)
INFO:tensorflow:lr = 9.065229e-05 (12.028 sec)
INFO:tensorflow:global_step/sec: 12.5065
INFO:tensorflow:loss = 1.2445109, step = 147800 (7.991 sec)
INFO:tensorflow:lr = 9.056123e-05 (7.991 sec)
INFO:tensorflow:global_step/sec: 12.7083
INFO:tensorflow:loss = 1.2437775, step = 147900 (7.869 sec)
INFO:tensorflow:lr = 9.047026e-05 (7.869 sec)
INFO:tensorflow:global_step/sec: 12.5151
INFO:tensorflow:loss = 1.247248, step = 148000 (7.991 sec)
INFO:tensorflow:lr = 9.0379384e-05 (7.992 sec)
INFO:tensorflow:global_step/sec: 12.2781
INFO:tensorflow:loss = 1.244056, step = 148100 (8.146 sec)
INFO:tensorflow:lr = 9.028859e-05 (8.147 sec)
INFO:tensorflow:global_step/sec: 12.6575
INFO:tensorflow:loss = 1.2453225, step = 148200 (7.899 sec)
INFO:tensorflow:lr = 9.019789e-05 (7.897 sec)
INFO:tensorflow:global_step/sec: 12.6457
INFO:tensorflow:loss = 1.2532479, step = 148300 (7.917 sec)
INFO:tensorflow:lr = 9.010729e-05 (7.918 sec)
INFO:tensorflow:global_step/sec: 12.6258
INFO:tensorflow:loss = 1.2450638, step = 148400 (7.911 sec)
INFO:tensorflow:lr = 9.001677e-05 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.4214
INFO:tensorflow:loss = 1.2449104, step = 148500 (8.051 sec)
INFO:tensorflow:lr = 8.9926354e-05 (8.050 sec)
INFO:tensorflow:global_step/sec: 12.9721
INFO:tensorflow:loss = 1.2434419, step = 148600 (7.709 sec)
INFO:tensorflow:lr = 8.9836016e-05 (7.717 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.28238
INFO:tensorflow:loss = 1.2493786, step = 148700 (12.079 sec)
INFO:tensorflow:lr = 8.9745765e-05 (12.071 sec)
INFO:tensorflow:global_step/sec: 12.7538
INFO:tensorflow:loss = 1.2464718, step = 148800 (7.841 sec)
INFO:tensorflow:lr = 8.9655616e-05 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.9883
INFO:tensorflow:loss = 1.2417631, step = 148900 (7.699 sec)
INFO:tensorflow:lr = 8.9565554e-05 (7.699 sec)
INFO:tensorflow:global_step/sec: 12.7568
INFO:tensorflow:loss = 1.2475207, step = 149000 (7.839 sec)
INFO:tensorflow:lr = 8.9475594e-05 (7.839 sec)
INFO:tensorflow:global_step/sec: 12.7225
INFO:tensorflow:loss = 1.245023, step = 149100 (7.860 sec)
INFO:tensorflow:lr = 8.938571e-05 (7.860 sec)
INFO:tensorflow:global_step/sec: 12.4462
INFO:tensorflow:loss = 1.245929, step = 149200 (8.034 sec)
INFO:tensorflow:lr = 8.9295914e-05 (8.034 sec)
INFO:tensorflow:global_step/sec: 12.679
INFO:tensorflow:loss = 1.2436159, step = 149300 (7.886 sec)
INFO:tensorflow:lr = 8.920622e-05 (7.886 sec)
INFO:tensorflow:global_step/sec: 12.6329
INFO:tensorflow:loss = 1.2462329, step = 149400 (7.917 sec)
INFO:tensorflow:lr = 8.91166e-05 (7.917 sec)
INFO:tensorflow:global_step/sec: 12.5809
INFO:tensorflow:loss = 1.2441921, step = 149500 (7.948 sec)
INFO:tensorflow:lr = 8.902709e-05 (7.949 sec)
INFO:tensorflow:global_step/sec: 12.7485
INFO:tensorflow:loss = 1.2438712, step = 149600 (7.840 sec)
INFO:tensorflow:lr = 8.893766e-05 (7.840 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.29717
INFO:tensorflow:loss = 1.2439181, step = 149700 (12.053 sec)
INFO:tensorflow:lr = 8.884831e-05 (12.054 sec)
INFO:tensorflow:global_step/sec: 12.469
INFO:tensorflow:loss = 1.2439133, step = 149800 (8.019 sec)
INFO:tensorflow:lr = 8.8759065e-05 (8.019 sec)
INFO:tensorflow:global_step/sec: 12.5088
INFO:tensorflow:loss = 1.243963, step = 149900 (7.994 sec)
INFO:tensorflow:lr = 8.86699e-05 (7.993 sec)
INFO:tensorflow:global_step/sec: 12.9688
INFO:tensorflow:loss = 1.2440338, step = 150000 (7.711 sec)
INFO:tensorflow:lr = 8.858084e-05 (7.711 sec)
INFO:tensorflow:global_step/sec: 12.8106
INFO:tensorflow:loss = 1.2429388, step = 150100 (7.806 sec)
INFO:tensorflow:lr = 8.849185e-05 (7.807 sec)
INFO:tensorflow:global_step/sec: 12.6214
INFO:tensorflow:loss = 1.2436142, step = 150200 (7.924 sec)
INFO:tensorflow:lr = 8.840295e-05 (7.923 sec)
INFO:tensorflow:global_step/sec: 12.281
INFO:tensorflow:loss = 1.2437512, step = 150300 (8.143 sec)
INFO:tensorflow:lr = 8.8314155e-05 (8.142 sec)
INFO:tensorflow:global_step/sec: 13.0256
INFO:tensorflow:loss = 1.2468607, step = 150400 (7.680 sec)
INFO:tensorflow:lr = 8.822544e-05 (7.681 sec)
INFO:tensorflow:global_step/sec: 12.1457
INFO:tensorflow:loss = 1.2475142, step = 150500 (8.236 sec)
INFO:tensorflow:lr = 8.8136825e-05 (8.236 sec)
INFO:tensorflow:global_step/sec: 12.1629
INFO:tensorflow:loss = 1.2439152, step = 150600 (8.227 sec)
INFO:tensorflow:lr = 8.8048284e-05 (8.227 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.27788
INFO:tensorflow:loss = 1.2460039, step = 150700 (12.070 sec)
INFO:tensorflow:lr = 8.795983e-05 (12.071 sec)
INFO:tensorflow:global_step/sec: 12.1396
INFO:tensorflow:loss = 1.244453, step = 150800 (8.235 sec)
INFO:tensorflow:lr = 8.7871485e-05 (8.236 sec)
INFO:tensorflow:global_step/sec: 12.3241
INFO:tensorflow:loss = 1.2413253, step = 150900 (8.114 sec)
INFO:tensorflow:lr = 8.7783206e-05 (8.113 sec)
INFO:tensorflow:global_step/sec: 12.7433
INFO:tensorflow:loss = 1.2435942, step = 151000 (7.848 sec)
INFO:tensorflow:lr = 8.769503e-05 (7.847 sec)
INFO:tensorflow:global_step/sec: 12.4367
INFO:tensorflow:loss = 1.2434167, step = 151100 (8.047 sec)
INFO:tensorflow:lr = 8.760694e-05 (8.048 sec)
INFO:tensorflow:global_step/sec: 12.1844
INFO:tensorflow:loss = 1.2427095, step = 151200 (8.200 sec)
INFO:tensorflow:lr = 8.751893e-05 (8.202 sec)
INFO:tensorflow:global_step/sec: 12.4128
INFO:tensorflow:loss = 1.2415925, step = 151300 (8.060 sec)
INFO:tensorflow:lr = 8.743102e-05 (8.057 sec)
INFO:tensorflow:global_step/sec: 12.3837
INFO:tensorflow:loss = 1.2425196, step = 151400 (8.078 sec)
INFO:tensorflow:lr = 8.7343185e-05 (8.078 sec)
INFO:tensorflow:global_step/sec: 12.7142
INFO:tensorflow:loss = 1.2500411, step = 151500 (7.865 sec)
INFO:tensorflow:lr = 8.725546e-05 (7.866 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.63353
INFO:tensorflow:loss = 1.2459517, step = 151600 (11.579 sec)
INFO:tensorflow:lr = 8.71678e-05 (11.582 sec)
INFO:tensorflow:global_step/sec: 12.2656
INFO:tensorflow:loss = 1.2423538, step = 151700 (8.149 sec)
INFO:tensorflow:lr = 8.708024e-05 (8.147 sec)
INFO:tensorflow:global_step/sec: 12.6121
INFO:tensorflow:loss = 1.2505404, step = 151800 (7.934 sec)
INFO:tensorflow:lr = 8.699277e-05 (7.934 sec)
INFO:tensorflow:global_step/sec: 12.7894
INFO:tensorflow:loss = 1.2456331, step = 151900 (7.814 sec)
INFO:tensorflow:lr = 8.690538e-05 (7.813 sec)
INFO:tensorflow:global_step/sec: 12.5493
INFO:tensorflow:loss = 1.246384, step = 152000 (7.971 sec)
INFO:tensorflow:lr = 8.681809e-05 (7.973 sec)
INFO:tensorflow:global_step/sec: 12.3397
INFO:tensorflow:loss = 1.2433354, step = 152100 (8.101 sec)
INFO:tensorflow:lr = 8.673086e-05 (8.100 sec)
INFO:tensorflow:global_step/sec: 12.3386
INFO:tensorflow:loss = 1.2442598, step = 152200 (8.106 sec)
INFO:tensorflow:lr = 8.664373e-05 (8.106 sec)
INFO:tensorflow:global_step/sec: 12.7453
INFO:tensorflow:loss = 1.2427703, step = 152300 (7.844 sec)
INFO:tensorflow:lr = 8.655672e-05 (7.847 sec)
INFO:tensorflow:global_step/sec: 12.9112
INFO:tensorflow:loss = 1.2528266, step = 152400 (7.746 sec)
INFO:tensorflow:lr = 8.6469765e-05 (7.744 sec)
INFO:tensorflow:global_step/sec: 12.8283
INFO:tensorflow:loss = 1.2462392, step = 152500 (7.801 sec)
INFO:tensorflow:lr = 8.6382905e-05 (7.800 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.06405
INFO:tensorflow:loss = 1.2444615, step = 152600 (11.031 sec)
INFO:tensorflow:lr = 8.6296124e-05 (11.031 sec)
INFO:tensorflow:global_step/sec: 13.1834
INFO:tensorflow:loss = 1.2434341, step = 152700 (7.585 sec)
INFO:tensorflow:lr = 8.6209424e-05 (7.585 sec)
INFO:tensorflow:global_step/sec: 12.618
INFO:tensorflow:loss = 1.2439532, step = 152800 (7.920 sec)
INFO:tensorflow:lr = 8.612283e-05 (7.921 sec)
INFO:tensorflow:global_step/sec: 12.5351
INFO:tensorflow:loss = 1.2488467, step = 152900 (7.978 sec)
INFO:tensorflow:lr = 8.6036314e-05 (7.978 sec)
INFO:tensorflow:global_step/sec: 12.4584
INFO:tensorflow:loss = 1.2502102, step = 153000 (8.026 sec)
INFO:tensorflow:lr = 8.59499e-05 (8.026 sec)
INFO:tensorflow:global_step/sec: 12.8334
INFO:tensorflow:loss = 1.2436447, step = 153100 (7.795 sec)
INFO:tensorflow:lr = 8.586356e-05 (7.798 sec)
INFO:tensorflow:global_step/sec: 12.9761
INFO:tensorflow:loss = 1.2446896, step = 153200 (7.709 sec)
INFO:tensorflow:lr = 8.5777305e-05 (7.707 sec)
INFO:tensorflow:global_step/sec: 12.5871
INFO:tensorflow:loss = 1.2448373, step = 153300 (7.939 sec)
INFO:tensorflow:lr = 8.569114e-05 (7.939 sec)
INFO:tensorflow:global_step/sec: 12.3519
INFO:tensorflow:loss = 1.24809, step = 153400 (8.097 sec)
INFO:tensorflow:lr = 8.560506e-05 (8.101 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1207
INFO:tensorflow:loss = 1.243688, step = 153500 (8.260 sec)
INFO:tensorflow:lr = 8.5519074e-05 (8.259 sec)
INFO:tensorflow:global_step/sec: 10.5819
INFO:tensorflow:loss = 1.2423753, step = 153600 (9.445 sec)
INFO:tensorflow:lr = 8.543317e-05 (9.445 sec)
INFO:tensorflow:global_step/sec: 12.5556
INFO:tensorflow:loss = 1.2469901, step = 153700 (7.965 sec)
INFO:tensorflow:lr = 8.534733e-05 (7.963 sec)
INFO:tensorflow:global_step/sec: 12.8293
INFO:tensorflow:loss = 1.2434926, step = 153800 (7.793 sec)
INFO:tensorflow:lr = 8.5261614e-05 (7.793 sec)
INFO:tensorflow:global_step/sec: 12.517
INFO:tensorflow:loss = 1.2455355, step = 153900 (7.987 sec)
INFO:tensorflow:lr = 8.517596e-05 (7.987 sec)
INFO:tensorflow:global_step/sec: 12.5719
INFO:tensorflow:loss = 1.2419705, step = 154000 (7.952 sec)
INFO:tensorflow:lr = 8.50904e-05 (7.954 sec)
INFO:tensorflow:global_step/sec: 12.5582
INFO:tensorflow:loss = 1.2438302, step = 154100 (7.966 sec)
INFO:tensorflow:lr = 8.500492e-05 (7.964 sec)
INFO:tensorflow:global_step/sec: 12.7461
INFO:tensorflow:loss = 1.2498709, step = 154200 (7.842 sec)
INFO:tensorflow:lr = 8.491953e-05 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.4289
INFO:tensorflow:loss = 1.2408761, step = 154300 (8.048 sec)
INFO:tensorflow:lr = 8.483423e-05 (8.048 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.6554
INFO:tensorflow:loss = 1.2410917, step = 154400 (7.906 sec)
INFO:tensorflow:lr = 8.474901e-05 (7.906 sec)
INFO:tensorflow:global_step/sec: 10.9855
INFO:tensorflow:loss = 1.2470044, step = 154500 (9.102 sec)
INFO:tensorflow:lr = 8.466389e-05 (9.103 sec)
INFO:tensorflow:global_step/sec: 12.8813
INFO:tensorflow:loss = 1.2450987, step = 154600 (7.757 sec)
INFO:tensorflow:lr = 8.457883e-05 (7.757 sec)
INFO:tensorflow:global_step/sec: 12.3704
INFO:tensorflow:loss = 1.243781, step = 154700 (8.091 sec)
INFO:tensorflow:lr = 8.449386e-05 (8.096 sec)
INFO:tensorflow:global_step/sec: 12.8307
INFO:tensorflow:loss = 1.2424403, step = 154800 (7.790 sec)
INFO:tensorflow:lr = 8.440899e-05 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.7739
INFO:tensorflow:loss = 1.2446392, step = 154900 (7.831 sec)
INFO:tensorflow:lr = 8.43242e-05 (7.829 sec)
INFO:tensorflow:global_step/sec: 12.4939
INFO:tensorflow:loss = 1.2413216, step = 155000 (7.998 sec)
INFO:tensorflow:lr = 8.42395e-05 (8.005 sec)
INFO:tensorflow:global_step/sec: 12.6264
INFO:tensorflow:loss = 1.2452991, step = 155100 (7.920 sec)
INFO:tensorflow:lr = 8.4154875e-05 (7.921 sec)
INFO:tensorflow:global_step/sec: 12.6364
INFO:tensorflow:loss = 1.2444293, step = 155200 (7.916 sec)
INFO:tensorflow:lr = 8.4070336e-05 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.6398
INFO:tensorflow:loss = 1.2430124, step = 155300 (7.916 sec)
INFO:tensorflow:lr = 8.398589e-05 (7.913 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0453
INFO:tensorflow:loss = 1.2424821, step = 155400 (8.309 sec)
INFO:tensorflow:lr = 8.390151e-05 (8.317 sec)
INFO:tensorflow:global_step/sec: 11.2988
INFO:tensorflow:loss = 1.2471951, step = 155500 (8.845 sec)
INFO:tensorflow:lr = 8.381725e-05 (8.837 sec)
INFO:tensorflow:global_step/sec: 12.8677
INFO:tensorflow:loss = 1.2544374, step = 155600 (7.768 sec)
INFO:tensorflow:lr = 8.3733044e-05 (7.768 sec)
INFO:tensorflow:global_step/sec: 12.6591
INFO:tensorflow:loss = 1.2498583, step = 155700 (7.899 sec)
INFO:tensorflow:lr = 8.364893e-05 (7.901 sec)
INFO:tensorflow:global_step/sec: 12.7241
INFO:tensorflow:loss = 1.2448832, step = 155800 (7.859 sec)
INFO:tensorflow:lr = 8.3564904e-05 (7.858 sec)
INFO:tensorflow:global_step/sec: 12.6629
INFO:tensorflow:loss = 1.244978, step = 155900 (7.894 sec)
INFO:tensorflow:lr = 8.348096e-05 (7.892 sec)
INFO:tensorflow:global_step/sec: 12.6465
INFO:tensorflow:loss = 1.2439777, step = 156000 (7.908 sec)
INFO:tensorflow:lr = 8.3397106e-05 (7.910 sec)
INFO:tensorflow:global_step/sec: 11.9459
INFO:tensorflow:loss = 1.2434794, step = 156100 (8.375 sec)
INFO:tensorflow:lr = 8.331332e-05 (8.374 sec)
INFO:tensorflow:global_step/sec: 12.5495
INFO:tensorflow:loss = 1.2426113, step = 156200 (7.969 sec)
INFO:tensorflow:lr = 8.322963e-05 (7.969 sec)
INFO:tensorflow:global_step/sec: 12.7922
INFO:tensorflow:loss = 1.2448606, step = 156300 (7.814 sec)
INFO:tensorflow:lr = 8.3146035e-05 (7.814 sec)
INFO:tensorflow:global_step/sec: 13.0554
INFO:tensorflow:loss = 1.2424155, step = 156400 (7.660 sec)
INFO:tensorflow:lr = 8.306251e-05 (7.660 sec)
INFO:tensorflow:Saving checkpoints for 156480 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2409595.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-156480
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-156480
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.731
INFO:tensorflow:Best (Exact Match) Accuracy: 0.731
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-156480
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 156480 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2545527, step = 156480
INFO:tensorflow:lr = 8.299575e-05
INFO:tensorflow:global_step/sec: 10.0548
INFO:tensorflow:loss = 1.2425964, step = 156580 (9.952 sec)
INFO:tensorflow:lr = 8.291238e-05 (9.953 sec)
INFO:tensorflow:global_step/sec: 12.7802
INFO:tensorflow:loss = 1.243327, step = 156680 (7.821 sec)
INFO:tensorflow:lr = 8.282909e-05 (7.821 sec)
INFO:tensorflow:global_step/sec: 12.544
INFO:tensorflow:loss = 1.2435379, step = 156780 (7.972 sec)
INFO:tensorflow:lr = 8.2745886e-05 (7.971 sec)
INFO:tensorflow:global_step/sec: 12.7211
INFO:tensorflow:loss = 1.2486848, step = 156880 (7.867 sec)
INFO:tensorflow:lr = 8.266277e-05 (7.867 sec)
INFO:tensorflow:global_step/sec: 13.0432
INFO:tensorflow:loss = 1.2423004, step = 156980 (7.664 sec)
INFO:tensorflow:lr = 8.2579725e-05 (7.663 sec)
INFO:tensorflow:global_step/sec: 12.7176
INFO:tensorflow:loss = 1.2446653, step = 157080 (7.861 sec)
INFO:tensorflow:lr = 8.249678e-05 (7.862 sec)
INFO:tensorflow:global_step/sec: 12.6615
INFO:tensorflow:loss = 1.2431146, step = 157180 (7.901 sec)
INFO:tensorflow:lr = 8.2413906e-05 (7.900 sec)
INFO:tensorflow:global_step/sec: 12.6109
INFO:tensorflow:loss = 1.2475405, step = 157280 (7.930 sec)
INFO:tensorflow:lr = 8.233111e-05 (7.930 sec)
INFO:tensorflow:global_step/sec: 12.9061
INFO:tensorflow:loss = 1.2426296, step = 157380 (7.748 sec)
INFO:tensorflow:lr = 8.2248414e-05 (7.748 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.17322
INFO:tensorflow:loss = 1.2472751, step = 157480 (12.237 sec)
INFO:tensorflow:lr = 8.2165796e-05 (12.237 sec)
INFO:tensorflow:global_step/sec: 12.8058
INFO:tensorflow:loss = 1.2430243, step = 157580 (7.808 sec)
INFO:tensorflow:lr = 8.208326e-05 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.7737
INFO:tensorflow:loss = 1.2482195, step = 157680 (7.823 sec)
INFO:tensorflow:lr = 8.20008e-05 (7.826 sec)
INFO:tensorflow:global_step/sec: 12.731
INFO:tensorflow:loss = 1.2453948, step = 157780 (7.857 sec)
INFO:tensorflow:lr = 8.191841e-05 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.7219
INFO:tensorflow:loss = 1.2427235, step = 157880 (7.866 sec)
INFO:tensorflow:lr = 8.183614e-05 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.5526
INFO:tensorflow:loss = 1.2457312, step = 157980 (7.964 sec)
INFO:tensorflow:lr = 8.1753926e-05 (7.964 sec)
INFO:tensorflow:global_step/sec: 12.6871
INFO:tensorflow:loss = 1.2447507, step = 158080 (7.878 sec)
INFO:tensorflow:lr = 8.167181e-05 (7.877 sec)
INFO:tensorflow:global_step/sec: 12.585
INFO:tensorflow:loss = 1.2416404, step = 158180 (7.946 sec)
INFO:tensorflow:lr = 8.1589766e-05 (7.949 sec)
INFO:tensorflow:global_step/sec: 12.7264
INFO:tensorflow:loss = 1.2412192, step = 158280 (7.863 sec)
INFO:tensorflow:lr = 8.15078e-05 (7.861 sec)
INFO:tensorflow:global_step/sec: 12.7235
INFO:tensorflow:loss = 1.2459588, step = 158380 (7.854 sec)
INFO:tensorflow:lr = 8.142593e-05 (7.854 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.34295
INFO:tensorflow:loss = 1.2420511, step = 158480 (11.987 sec)
INFO:tensorflow:lr = 8.134413e-05 (11.986 sec)
INFO:tensorflow:global_step/sec: 12.714
INFO:tensorflow:loss = 1.2457771, step = 158580 (7.865 sec)
INFO:tensorflow:lr = 8.126243e-05 (7.864 sec)
INFO:tensorflow:global_step/sec: 12.6987
INFO:tensorflow:loss = 1.2400635, step = 158680 (7.875 sec)
INFO:tensorflow:lr = 8.11808e-05 (7.875 sec)
INFO:tensorflow:global_step/sec: 12.8282
INFO:tensorflow:loss = 1.2438774, step = 158780 (7.796 sec)
INFO:tensorflow:lr = 8.109924e-05 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.8694
INFO:tensorflow:loss = 1.2427423, step = 158880 (7.770 sec)
INFO:tensorflow:lr = 8.101778e-05 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.6658
INFO:tensorflow:loss = 1.241447, step = 158980 (7.896 sec)
INFO:tensorflow:lr = 8.093639e-05 (7.896 sec)
INFO:tensorflow:global_step/sec: 12.5078
INFO:tensorflow:loss = 1.2422627, step = 159080 (7.998 sec)
INFO:tensorflow:lr = 8.08551e-05 (8.001 sec)
INFO:tensorflow:global_step/sec: 12.6664
INFO:tensorflow:loss = 1.250941, step = 159180 (7.897 sec)
INFO:tensorflow:lr = 8.077386e-05 (7.900 sec)
INFO:tensorflow:global_step/sec: 12.7541
INFO:tensorflow:loss = 1.2464032, step = 159280 (7.841 sec)
INFO:tensorflow:lr = 8.069272e-05 (7.836 sec)
INFO:tensorflow:global_step/sec: 12.7697
INFO:tensorflow:loss = 1.2432774, step = 159380 (7.831 sec)
INFO:tensorflow:lr = 8.0611666e-05 (7.831 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.30956
INFO:tensorflow:loss = 1.245586, step = 159480 (12.031 sec)
INFO:tensorflow:lr = 8.053069e-05 (12.032 sec)
INFO:tensorflow:global_step/sec: 12.7577
INFO:tensorflow:loss = 1.2436898, step = 159580 (7.840 sec)
INFO:tensorflow:lr = 8.0449805e-05 (7.840 sec)
INFO:tensorflow:global_step/sec: 12.6921
INFO:tensorflow:loss = 1.2427038, step = 159680 (7.880 sec)
INFO:tensorflow:lr = 8.036898e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.5858
INFO:tensorflow:loss = 1.2463199, step = 159780 (7.940 sec)
INFO:tensorflow:lr = 8.028825e-05 (7.940 sec)
INFO:tensorflow:global_step/sec: 12.9176
INFO:tensorflow:loss = 1.2422549, step = 159880 (7.743 sec)
INFO:tensorflow:lr = 8.02076e-05 (7.746 sec)
INFO:tensorflow:global_step/sec: 12.7488
INFO:tensorflow:loss = 1.2433963, step = 159980 (7.847 sec)
INFO:tensorflow:lr = 8.0127036e-05 (7.844 sec)
INFO:tensorflow:global_step/sec: 12.8153
INFO:tensorflow:loss = 1.2405639, step = 160080 (7.804 sec)
INFO:tensorflow:lr = 8.004654e-05 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.7982
INFO:tensorflow:loss = 1.2429812, step = 160180 (7.810 sec)
INFO:tensorflow:lr = 7.996613e-05 (7.810 sec)
INFO:tensorflow:global_step/sec: 12.5992
INFO:tensorflow:loss = 1.2426969, step = 160280 (7.935 sec)
INFO:tensorflow:lr = 7.9885795e-05 (7.935 sec)
INFO:tensorflow:global_step/sec: 12.6284
INFO:tensorflow:loss = 1.2485273, step = 160380 (7.924 sec)
INFO:tensorflow:lr = 7.9805555e-05 (7.924 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.43014
INFO:tensorflow:loss = 1.2451617, step = 160480 (11.859 sec)
INFO:tensorflow:lr = 7.972538e-05 (11.859 sec)
INFO:tensorflow:global_step/sec: 12.8098
INFO:tensorflow:loss = 1.2438232, step = 160580 (7.804 sec)
INFO:tensorflow:lr = 7.9645295e-05 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.8416
INFO:tensorflow:loss = 1.2421925, step = 160680 (7.787 sec)
INFO:tensorflow:lr = 7.9565296e-05 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.8333
INFO:tensorflow:loss = 1.2429509, step = 160780 (7.792 sec)
INFO:tensorflow:lr = 7.948537e-05 (7.792 sec)
INFO:tensorflow:global_step/sec: 12.7757
INFO:tensorflow:loss = 1.2418758, step = 160880 (7.828 sec)
INFO:tensorflow:lr = 7.940552e-05 (7.828 sec)
INFO:tensorflow:global_step/sec: 12.6713
INFO:tensorflow:loss = 1.2432858, step = 160980 (7.896 sec)
INFO:tensorflow:lr = 7.9325764e-05 (7.896 sec)
INFO:tensorflow:global_step/sec: 12.5948
INFO:tensorflow:loss = 1.24457, step = 161080 (7.943 sec)
INFO:tensorflow:lr = 7.9246085e-05 (7.944 sec)
INFO:tensorflow:global_step/sec: 12.7125
INFO:tensorflow:loss = 1.2471696, step = 161180 (7.864 sec)
INFO:tensorflow:lr = 7.916647e-05 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.8499
INFO:tensorflow:loss = 1.2433884, step = 161280 (7.782 sec)
INFO:tensorflow:lr = 7.9086945e-05 (7.782 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.62131
INFO:tensorflow:loss = 1.2433494, step = 161380 (11.599 sec)
INFO:tensorflow:lr = 7.900751e-05 (11.598 sec)
INFO:tensorflow:global_step/sec: 12.575
INFO:tensorflow:loss = 1.2430332, step = 161480 (7.952 sec)
INFO:tensorflow:lr = 7.892813e-05 (7.952 sec)
INFO:tensorflow:global_step/sec: 12.7951
INFO:tensorflow:loss = 1.2427179, step = 161580 (7.815 sec)
INFO:tensorflow:lr = 7.884885e-05 (7.815 sec)
INFO:tensorflow:global_step/sec: 12.7562
INFO:tensorflow:loss = 1.2430513, step = 161680 (7.841 sec)
INFO:tensorflow:lr = 7.876965e-05 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.3947
INFO:tensorflow:loss = 1.2428813, step = 161780 (8.065 sec)
INFO:tensorflow:lr = 7.8690515e-05 (8.064 sec)
INFO:tensorflow:global_step/sec: 12.5447
INFO:tensorflow:loss = 1.2428411, step = 161880 (7.968 sec)
INFO:tensorflow:lr = 7.8611476e-05 (7.969 sec)
INFO:tensorflow:global_step/sec: 12.5711
INFO:tensorflow:loss = 1.2425469, step = 161980 (7.958 sec)
INFO:tensorflow:lr = 7.8532496e-05 (7.958 sec)
INFO:tensorflow:global_step/sec: 12.9131
INFO:tensorflow:loss = 1.2412182, step = 162080 (7.746 sec)
INFO:tensorflow:lr = 7.845362e-05 (7.746 sec)
INFO:tensorflow:global_step/sec: 12.4472
INFO:tensorflow:loss = 1.2435589, step = 162180 (8.033 sec)
INFO:tensorflow:lr = 7.8374804e-05 (8.034 sec)
INFO:tensorflow:global_step/sec: 12.8406
INFO:tensorflow:loss = 1.2414767, step = 162280 (7.783 sec)
INFO:tensorflow:lr = 7.829606e-05 (7.784 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.14022
INFO:tensorflow:loss = 1.242339, step = 162380 (10.944 sec)
INFO:tensorflow:lr = 7.8217425e-05 (10.944 sec)
INFO:tensorflow:global_step/sec: 12.7715
INFO:tensorflow:loss = 1.2425467, step = 162480 (7.827 sec)
INFO:tensorflow:lr = 7.8138844e-05 (7.828 sec)
INFO:tensorflow:global_step/sec: 12.3174
INFO:tensorflow:loss = 1.2458082, step = 162580 (8.123 sec)
INFO:tensorflow:lr = 7.8060366e-05 (8.122 sec)
INFO:tensorflow:global_step/sec: 12.43
INFO:tensorflow:loss = 1.2450854, step = 162680 (8.040 sec)
INFO:tensorflow:lr = 7.7981946e-05 (8.040 sec)
INFO:tensorflow:global_step/sec: 12.8185
INFO:tensorflow:loss = 1.2413714, step = 162780 (7.808 sec)
INFO:tensorflow:lr = 7.7903605e-05 (7.808 sec)
INFO:tensorflow:global_step/sec: 12.8459
INFO:tensorflow:loss = 1.2418839, step = 162880 (7.782 sec)
INFO:tensorflow:lr = 7.782536e-05 (7.782 sec)
INFO:tensorflow:global_step/sec: 12.6585
INFO:tensorflow:loss = 1.2452873, step = 162980 (7.896 sec)
INFO:tensorflow:lr = 7.774718e-05 (7.896 sec)
INFO:tensorflow:global_step/sec: 12.9969
INFO:tensorflow:loss = 1.2421685, step = 163080 (7.699 sec)
INFO:tensorflow:lr = 7.766908e-05 (7.698 sec)
INFO:tensorflow:global_step/sec: 12.4813
INFO:tensorflow:loss = 1.2429278, step = 163180 (8.012 sec)
INFO:tensorflow:lr = 7.759106e-05 (8.012 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2018
INFO:tensorflow:loss = 1.2407615, step = 163280 (8.198 sec)
INFO:tensorflow:lr = 7.751311e-05 (8.200 sec)
INFO:tensorflow:global_step/sec: 10.6203
INFO:tensorflow:loss = 1.2463655, step = 163380 (9.414 sec)
INFO:tensorflow:lr = 7.743525e-05 (9.412 sec)
INFO:tensorflow:global_step/sec: 12.751
INFO:tensorflow:loss = 1.2475263, step = 163480 (7.838 sec)
INFO:tensorflow:lr = 7.7357465e-05 (7.840 sec)
INFO:tensorflow:global_step/sec: 12.8988
INFO:tensorflow:loss = 1.2489913, step = 163580 (7.759 sec)
INFO:tensorflow:lr = 7.7279765e-05 (7.758 sec)
INFO:tensorflow:global_step/sec: 12.7432
INFO:tensorflow:loss = 1.246935, step = 163680 (7.840 sec)
INFO:tensorflow:lr = 7.720213e-05 (7.840 sec)
INFO:tensorflow:global_step/sec: 12.7445
INFO:tensorflow:loss = 1.2567366, step = 163780 (7.846 sec)
INFO:tensorflow:lr = 7.7124576e-05 (7.852 sec)
INFO:tensorflow:global_step/sec: 12.6891
INFO:tensorflow:loss = 1.2523156, step = 163880 (7.886 sec)
INFO:tensorflow:lr = 7.704711e-05 (7.881 sec)
INFO:tensorflow:global_step/sec: 12.5144
INFO:tensorflow:loss = 1.2429801, step = 163980 (7.989 sec)
INFO:tensorflow:lr = 7.696971e-05 (7.989 sec)
INFO:tensorflow:global_step/sec: 12.7412
INFO:tensorflow:loss = 1.2436205, step = 164080 (7.852 sec)
INFO:tensorflow:lr = 7.689239e-05 (7.851 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.3508
INFO:tensorflow:loss = 1.2439187, step = 164180 (8.093 sec)
INFO:tensorflow:lr = 7.681516e-05 (8.093 sec)
INFO:tensorflow:global_step/sec: 10.8155
INFO:tensorflow:loss = 1.2418324, step = 164280 (9.247 sec)
INFO:tensorflow:lr = 7.673798e-05 (9.246 sec)
INFO:tensorflow:global_step/sec: 12.8791
INFO:tensorflow:loss = 1.2434556, step = 164380 (7.769 sec)
INFO:tensorflow:lr = 7.66609e-05 (7.770 sec)
INFO:tensorflow:global_step/sec: 12.7672
INFO:tensorflow:loss = 1.2411584, step = 164480 (7.824 sec)
INFO:tensorflow:lr = 7.658389e-05 (7.826 sec)
INFO:tensorflow:global_step/sec: 12.8412
INFO:tensorflow:loss = 1.243303, step = 164580 (7.787 sec)
INFO:tensorflow:lr = 7.650696e-05 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.6858
INFO:tensorflow:loss = 1.243237, step = 164680 (7.883 sec)
INFO:tensorflow:lr = 7.643011e-05 (7.882 sec)
INFO:tensorflow:global_step/sec: 12.7993
INFO:tensorflow:loss = 1.2468959, step = 164780 (7.819 sec)
INFO:tensorflow:lr = 7.6353324e-05 (7.819 sec)
INFO:tensorflow:global_step/sec: 12.9195
INFO:tensorflow:loss = 1.2466041, step = 164880 (7.735 sec)
INFO:tensorflow:lr = 7.627664e-05 (7.735 sec)
INFO:tensorflow:global_step/sec: 12.6698
INFO:tensorflow:loss = 1.2453058, step = 164980 (7.897 sec)
INFO:tensorflow:lr = 7.6200005e-05 (7.896 sec)
INFO:tensorflow:global_step/sec: 12.608
INFO:tensorflow:loss = 1.2435452, step = 165080 (7.932 sec)
INFO:tensorflow:lr = 7.612347e-05 (7.932 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.9524
INFO:tensorflow:loss = 1.2424791, step = 165180 (8.366 sec)
INFO:tensorflow:lr = 7.604699e-05 (8.367 sec)
INFO:tensorflow:global_step/sec: 11.3109
INFO:tensorflow:loss = 1.2438216, step = 165280 (8.836 sec)
INFO:tensorflow:lr = 7.597061e-05 (8.835 sec)
INFO:tensorflow:global_step/sec: 13.0587
INFO:tensorflow:loss = 1.2450823, step = 165380 (7.660 sec)
INFO:tensorflow:lr = 7.58943e-05 (7.662 sec)
INFO:tensorflow:global_step/sec: 12.7891
INFO:tensorflow:loss = 1.2420415, step = 165480 (7.818 sec)
INFO:tensorflow:lr = 7.5818054e-05 (7.821 sec)
INFO:tensorflow:global_step/sec: 12.3887
INFO:tensorflow:loss = 1.2473958, step = 165580 (8.077 sec)
INFO:tensorflow:lr = 7.574189e-05 (8.072 sec)
INFO:tensorflow:global_step/sec: 12.3917
INFO:tensorflow:loss = 1.2471696, step = 165680 (8.069 sec)
INFO:tensorflow:lr = 7.566581e-05 (8.070 sec)
INFO:tensorflow:global_step/sec: 12.9346
INFO:tensorflow:loss = 1.2463306, step = 165780 (7.729 sec)
INFO:tensorflow:lr = 7.558979e-05 (7.728 sec)
INFO:tensorflow:global_step/sec: 12.6382
INFO:tensorflow:loss = 1.2409977, step = 165880 (7.917 sec)
INFO:tensorflow:lr = 7.551387e-05 (7.916 sec)
INFO:tensorflow:global_step/sec: 12.874
INFO:tensorflow:loss = 1.2447242, step = 165980 (7.764 sec)
INFO:tensorflow:lr = 7.5438e-05 (7.765 sec)
INFO:tensorflow:global_step/sec: 12.5758
INFO:tensorflow:loss = 1.2421839, step = 166080 (7.948 sec)
INFO:tensorflow:lr = 7.536224e-05 (7.948 sec)
INFO:tensorflow:global_step/sec: 12.8224
INFO:tensorflow:loss = 1.2433058, step = 166180 (7.799 sec)
INFO:tensorflow:lr = 7.528653e-05 (7.801 sec)
INFO:tensorflow:Saving checkpoints for 166260 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2428299.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-166260
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-166260
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.731
INFO:tensorflow:Best (Exact Match) Accuracy: 0.731
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-166260
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 166260 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2467928, step = 166260
INFO:tensorflow:lr = 7.522602e-05
INFO:tensorflow:global_step/sec: 9.96783
INFO:tensorflow:loss = 1.2433182, step = 166360 (10.039 sec)
INFO:tensorflow:lr = 7.5150456e-05 (10.038 sec)
INFO:tensorflow:global_step/sec: 12.6081
INFO:tensorflow:loss = 1.2423993, step = 166460 (7.930 sec)
INFO:tensorflow:lr = 7.507496e-05 (7.930 sec)
INFO:tensorflow:global_step/sec: 12.9082
INFO:tensorflow:loss = 1.2489258, step = 166560 (7.745 sec)
INFO:tensorflow:lr = 7.499954e-05 (7.745 sec)
INFO:tensorflow:global_step/sec: 12.7743
INFO:tensorflow:loss = 1.2423285, step = 166660 (7.827 sec)
INFO:tensorflow:lr = 7.492421e-05 (7.828 sec)
INFO:tensorflow:global_step/sec: 12.5581
INFO:tensorflow:loss = 1.2450229, step = 166760 (7.962 sec)
INFO:tensorflow:lr = 7.484895e-05 (7.961 sec)
INFO:tensorflow:global_step/sec: 12.8234
INFO:tensorflow:loss = 1.2427442, step = 166860 (7.804 sec)
INFO:tensorflow:lr = 7.4773765e-05 (7.805 sec)
INFO:tensorflow:global_step/sec: 12.8935
INFO:tensorflow:loss = 1.243734, step = 166960 (7.754 sec)
INFO:tensorflow:lr = 7.469865e-05 (7.753 sec)
INFO:tensorflow:global_step/sec: 13.0232
INFO:tensorflow:loss = 1.243697, step = 167060 (7.676 sec)
INFO:tensorflow:lr = 7.4623604e-05 (7.675 sec)
INFO:tensorflow:global_step/sec: 12.8317
INFO:tensorflow:loss = 1.2420083, step = 167160 (7.793 sec)
INFO:tensorflow:lr = 7.454865e-05 (7.794 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.19419
INFO:tensorflow:loss = 1.2424514, step = 167260 (12.208 sec)
INFO:tensorflow:lr = 7.447377e-05 (12.208 sec)
INFO:tensorflow:global_step/sec: 12.8685
INFO:tensorflow:loss = 1.2442771, step = 167360 (7.766 sec)
INFO:tensorflow:lr = 7.4398966e-05 (7.768 sec)
INFO:tensorflow:global_step/sec: 12.7135
INFO:tensorflow:loss = 1.2463063, step = 167460 (7.866 sec)
INFO:tensorflow:lr = 7.432422e-05 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.8504
INFO:tensorflow:loss = 1.2478987, step = 167560 (7.782 sec)
INFO:tensorflow:lr = 7.424955e-05 (7.784 sec)
INFO:tensorflow:global_step/sec: 12.4762
INFO:tensorflow:loss = 1.2463549, step = 167660 (8.017 sec)
INFO:tensorflow:lr = 7.4174975e-05 (8.014 sec)
INFO:tensorflow:global_step/sec: 12.6068
INFO:tensorflow:loss = 1.243028, step = 167760 (7.935 sec)
INFO:tensorflow:lr = 7.4100455e-05 (7.936 sec)
INFO:tensorflow:global_step/sec: 12.7757
INFO:tensorflow:loss = 1.2428224, step = 167860 (7.827 sec)
INFO:tensorflow:lr = 7.402602e-05 (7.827 sec)
INFO:tensorflow:global_step/sec: 12.9328
INFO:tensorflow:loss = 1.2443757, step = 167960 (7.733 sec)
INFO:tensorflow:lr = 7.3951654e-05 (7.732 sec)
INFO:tensorflow:global_step/sec: 12.7784
INFO:tensorflow:loss = 1.242498, step = 168060 (7.827 sec)
INFO:tensorflow:lr = 7.387737e-05 (7.827 sec)
INFO:tensorflow:global_step/sec: 12.7597
INFO:tensorflow:loss = 1.246404, step = 168160 (7.835 sec)
INFO:tensorflow:lr = 7.380317e-05 (7.837 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.28793
INFO:tensorflow:loss = 1.2430749, step = 168260 (12.061 sec)
INFO:tensorflow:lr = 7.3729025e-05 (12.062 sec)
INFO:tensorflow:global_step/sec: 12.7741
INFO:tensorflow:loss = 1.2416666, step = 168360 (7.837 sec)
INFO:tensorflow:lr = 7.365496e-05 (7.835 sec)
INFO:tensorflow:global_step/sec: 12.6338
INFO:tensorflow:loss = 1.2404863, step = 168460 (7.911 sec)
INFO:tensorflow:lr = 7.358097e-05 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.8762
INFO:tensorflow:loss = 1.2460499, step = 168560 (7.765 sec)
INFO:tensorflow:lr = 7.3507064e-05 (7.767 sec)
INFO:tensorflow:global_step/sec: 12.6483
INFO:tensorflow:loss = 1.245742, step = 168660 (7.905 sec)
INFO:tensorflow:lr = 7.343323e-05 (7.904 sec)
INFO:tensorflow:global_step/sec: 12.7735
INFO:tensorflow:loss = 1.2437085, step = 168760 (7.827 sec)
INFO:tensorflow:lr = 7.335946e-05 (7.828 sec)
INFO:tensorflow:global_step/sec: 12.7082
INFO:tensorflow:loss = 1.2422924, step = 168860 (7.873 sec)
INFO:tensorflow:lr = 7.3285766e-05 (7.873 sec)
INFO:tensorflow:global_step/sec: 12.9294
INFO:tensorflow:loss = 1.2445481, step = 168960 (7.737 sec)
INFO:tensorflow:lr = 7.321215e-05 (7.736 sec)
INFO:tensorflow:global_step/sec: 13.0217
INFO:tensorflow:loss = 1.2428153, step = 169060 (7.677 sec)
INFO:tensorflow:lr = 7.3138595e-05 (7.679 sec)
INFO:tensorflow:global_step/sec: 12.4694
INFO:tensorflow:loss = 1.24292, step = 169160 (8.022 sec)
INFO:tensorflow:lr = 7.306513e-05 (8.020 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.20923
INFO:tensorflow:loss = 1.2442397, step = 169260 (12.182 sec)
INFO:tensorflow:lr = 7.299173e-05 (12.182 sec)
INFO:tensorflow:global_step/sec: 12.6733
INFO:tensorflow:loss = 1.2442619, step = 169360 (7.886 sec)
INFO:tensorflow:lr = 7.291842e-05 (7.887 sec)
INFO:tensorflow:global_step/sec: 12.714
INFO:tensorflow:loss = 1.2440118, step = 169460 (7.871 sec)
INFO:tensorflow:lr = 7.284516e-05 (7.870 sec)
INFO:tensorflow:global_step/sec: 12.5632
INFO:tensorflow:loss = 1.2449383, step = 169560 (7.957 sec)
INFO:tensorflow:lr = 7.277198e-05 (7.959 sec)
INFO:tensorflow:global_step/sec: 12.7235
INFO:tensorflow:loss = 1.2519269, step = 169660 (7.862 sec)
INFO:tensorflow:lr = 7.269889e-05 (7.861 sec)
INFO:tensorflow:global_step/sec: 12.8846
INFO:tensorflow:loss = 1.2426889, step = 169760 (7.758 sec)
INFO:tensorflow:lr = 7.262586e-05 (7.759 sec)
INFO:tensorflow:global_step/sec: 12.6476
INFO:tensorflow:loss = 1.2445718, step = 169860 (7.902 sec)
INFO:tensorflow:lr = 7.25529e-05 (7.902 sec)
INFO:tensorflow:global_step/sec: 12.6015
INFO:tensorflow:loss = 1.242896, step = 169960 (7.939 sec)
INFO:tensorflow:lr = 7.248003e-05 (7.941 sec)
INFO:tensorflow:global_step/sec: 12.5507
INFO:tensorflow:loss = 1.2418842, step = 170060 (7.964 sec)
INFO:tensorflow:lr = 7.2407216e-05 (7.962 sec)
INFO:tensorflow:global_step/sec: 13.0502
INFO:tensorflow:loss = 1.2443966, step = 170160 (7.668 sec)
INFO:tensorflow:lr = 7.2334486e-05 (7.668 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.36448
INFO:tensorflow:loss = 1.2439117, step = 170260 (11.952 sec)
INFO:tensorflow:lr = 7.226182e-05 (11.954 sec)
INFO:tensorflow:global_step/sec: 12.879
INFO:tensorflow:loss = 1.2425317, step = 170360 (7.767 sec)
INFO:tensorflow:lr = 7.2189236e-05 (7.764 sec)
INFO:tensorflow:global_step/sec: 13.0269
INFO:tensorflow:loss = 1.243981, step = 170460 (7.677 sec)
INFO:tensorflow:lr = 7.2116716e-05 (7.677 sec)
INFO:tensorflow:global_step/sec: 12.7859
INFO:tensorflow:loss = 1.2428844, step = 170560 (7.824 sec)
INFO:tensorflow:lr = 7.204427e-05 (7.824 sec)
INFO:tensorflow:global_step/sec: 12.6076
INFO:tensorflow:loss = 1.2415282, step = 170660 (7.926 sec)
INFO:tensorflow:lr = 7.19719e-05 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.8436
INFO:tensorflow:loss = 1.242303, step = 170760 (7.788 sec)
INFO:tensorflow:lr = 7.189961e-05 (7.787 sec)
INFO:tensorflow:global_step/sec: 12.3776
INFO:tensorflow:loss = 1.2428727, step = 170860 (8.078 sec)
INFO:tensorflow:lr = 7.182739e-05 (8.080 sec)
INFO:tensorflow:global_step/sec: 12.614
INFO:tensorflow:loss = 1.245311, step = 170960 (7.930 sec)
INFO:tensorflow:lr = 7.1755225e-05 (7.928 sec)
INFO:tensorflow:global_step/sec: 12.8587
INFO:tensorflow:loss = 1.2429055, step = 171060 (7.771 sec)
INFO:tensorflow:lr = 7.1683135e-05 (7.774 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.60485
INFO:tensorflow:loss = 1.2450764, step = 171160 (11.623 sec)
INFO:tensorflow:lr = 7.161113e-05 (11.622 sec)
INFO:tensorflow:global_step/sec: 12.5893
INFO:tensorflow:loss = 1.2422101, step = 171260 (7.941 sec)
INFO:tensorflow:lr = 7.15392e-05 (7.941 sec)
INFO:tensorflow:global_step/sec: 12.1946
INFO:tensorflow:loss = 1.2442441, step = 171360 (8.204 sec)
INFO:tensorflow:lr = 7.1467344e-05 (8.203 sec)
INFO:tensorflow:global_step/sec: 12.5814
INFO:tensorflow:loss = 1.2420107, step = 171460 (7.948 sec)
INFO:tensorflow:lr = 7.139555e-05 (7.950 sec)
INFO:tensorflow:global_step/sec: 12.8204
INFO:tensorflow:loss = 1.2425112, step = 171560 (7.802 sec)
INFO:tensorflow:lr = 7.1323826e-05 (7.801 sec)
INFO:tensorflow:global_step/sec: 12.731
INFO:tensorflow:loss = 1.2440164, step = 171660 (7.851 sec)
INFO:tensorflow:lr = 7.125219e-05 (7.854 sec)
INFO:tensorflow:global_step/sec: 13.0197
INFO:tensorflow:loss = 1.2427493, step = 171760 (7.682 sec)
INFO:tensorflow:lr = 7.118061e-05 (7.678 sec)
INFO:tensorflow:global_step/sec: 12.7282
INFO:tensorflow:loss = 1.2409182, step = 171860 (7.854 sec)
INFO:tensorflow:lr = 7.1109105e-05 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.8746
INFO:tensorflow:loss = 1.2442806, step = 171960 (7.767 sec)
INFO:tensorflow:lr = 7.103767e-05 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.542
INFO:tensorflow:loss = 1.2414162, step = 172060 (7.973 sec)
INFO:tensorflow:lr = 7.0966315e-05 (7.971 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.09987
INFO:tensorflow:loss = 1.2424979, step = 172160 (10.990 sec)
INFO:tensorflow:lr = 7.089503e-05 (10.990 sec)
INFO:tensorflow:global_step/sec: 12.881
INFO:tensorflow:loss = 1.255378, step = 172260 (7.762 sec)
INFO:tensorflow:lr = 7.082381e-05 (7.762 sec)
INFO:tensorflow:global_step/sec: 12.6186
INFO:tensorflow:loss = 1.2428926, step = 172360 (7.925 sec)
INFO:tensorflow:lr = 7.075266e-05 (7.925 sec)
INFO:tensorflow:global_step/sec: 12.5197
INFO:tensorflow:loss = 1.2457323, step = 172460 (7.988 sec)
INFO:tensorflow:lr = 7.068159e-05 (7.988 sec)
INFO:tensorflow:global_step/sec: 13.1134
INFO:tensorflow:loss = 1.2408348, step = 172560 (7.625 sec)
INFO:tensorflow:lr = 7.0610586e-05 (7.625 sec)
INFO:tensorflow:global_step/sec: 12.8617
INFO:tensorflow:loss = 1.242487, step = 172660 (7.780 sec)
INFO:tensorflow:lr = 7.053967e-05 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.6416
INFO:tensorflow:loss = 1.2421248, step = 172760 (7.911 sec)
INFO:tensorflow:lr = 7.04688e-05 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.5986
INFO:tensorflow:loss = 1.240845, step = 172860 (7.934 sec)
INFO:tensorflow:lr = 7.0398026e-05 (7.934 sec)
INFO:tensorflow:global_step/sec: 12.9582
INFO:tensorflow:loss = 1.2410495, step = 172960 (7.721 sec)
INFO:tensorflow:lr = 7.0327296e-05 (7.721 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0786
INFO:tensorflow:loss = 1.2426779, step = 173060 (8.279 sec)
INFO:tensorflow:lr = 7.025665e-05 (8.280 sec)
INFO:tensorflow:global_step/sec: 10.6016
INFO:tensorflow:loss = 1.2406965, step = 173160 (9.433 sec)
INFO:tensorflow:lr = 7.0186084e-05 (9.432 sec)
INFO:tensorflow:global_step/sec: 12.9924
INFO:tensorflow:loss = 1.2419554, step = 173260 (7.695 sec)
INFO:tensorflow:lr = 7.0115566e-05 (7.697 sec)
INFO:tensorflow:global_step/sec: 12.6805
INFO:tensorflow:loss = 1.2415868, step = 173360 (7.886 sec)
INFO:tensorflow:lr = 7.004514e-05 (7.886 sec)
INFO:tensorflow:global_step/sec: 12.8956
INFO:tensorflow:loss = 1.2414397, step = 173460 (7.751 sec)
INFO:tensorflow:lr = 6.9974776e-05 (7.750 sec)
INFO:tensorflow:global_step/sec: 12.8402
INFO:tensorflow:loss = 1.2436533, step = 173560 (7.790 sec)
INFO:tensorflow:lr = 6.9904476e-05 (7.789 sec)
INFO:tensorflow:global_step/sec: 12.6078
INFO:tensorflow:loss = 1.2412962, step = 173660 (7.932 sec)
INFO:tensorflow:lr = 6.983426e-05 (7.932 sec)
INFO:tensorflow:global_step/sec: 12.9803
INFO:tensorflow:loss = 1.2563962, step = 173760 (7.706 sec)
INFO:tensorflow:lr = 6.976411e-05 (7.708 sec)
INFO:tensorflow:global_step/sec: 12.5698
INFO:tensorflow:loss = 1.2463813, step = 173860 (7.951 sec)
INFO:tensorflow:lr = 6.9694026e-05 (7.949 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.4643
INFO:tensorflow:loss = 1.2427906, step = 173960 (8.028 sec)
INFO:tensorflow:lr = 6.9624024e-05 (8.030 sec)
INFO:tensorflow:global_step/sec: 11.041
INFO:tensorflow:loss = 1.2439783, step = 174060 (9.053 sec)
INFO:tensorflow:lr = 6.955408e-05 (9.055 sec)
INFO:tensorflow:global_step/sec: 12.7311
INFO:tensorflow:loss = 1.2426528, step = 174160 (7.855 sec)
INFO:tensorflow:lr = 6.9484224e-05 (7.853 sec)
INFO:tensorflow:global_step/sec: 12.6687
INFO:tensorflow:loss = 1.2417759, step = 174260 (7.896 sec)
INFO:tensorflow:lr = 6.941441e-05 (7.895 sec)
INFO:tensorflow:global_step/sec: 12.5623
INFO:tensorflow:loss = 1.2435216, step = 174360 (7.957 sec)
INFO:tensorflow:lr = 6.9344685e-05 (7.958 sec)
INFO:tensorflow:global_step/sec: 12.637
INFO:tensorflow:loss = 1.2440283, step = 174460 (7.913 sec)
INFO:tensorflow:lr = 6.927503e-05 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.9613
INFO:tensorflow:loss = 1.2432188, step = 174560 (7.722 sec)
INFO:tensorflow:lr = 6.920543e-05 (7.721 sec)
INFO:tensorflow:global_step/sec: 12.822
INFO:tensorflow:loss = 1.2439592, step = 174660 (7.798 sec)
INFO:tensorflow:lr = 6.9135924e-05 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.7043
INFO:tensorflow:loss = 1.2439635, step = 174760 (7.866 sec)
INFO:tensorflow:lr = 6.9066475e-05 (7.868 sec)
INFO:tensorflow:global_step/sec: 12.64
INFO:tensorflow:loss = 1.2470303, step = 174860 (7.913 sec)
INFO:tensorflow:lr = 6.899709e-05 (7.911 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.29
INFO:tensorflow:loss = 1.242181, step = 174960 (8.140 sec)
INFO:tensorflow:lr = 6.892779e-05 (8.147 sec)
INFO:tensorflow:global_step/sec: 11.6222
INFO:tensorflow:loss = 1.2408583, step = 175060 (8.603 sec)
INFO:tensorflow:lr = 6.8858535e-05 (8.596 sec)
INFO:tensorflow:global_step/sec: 12.5978
INFO:tensorflow:loss = 1.2463039, step = 175160 (7.941 sec)
INFO:tensorflow:lr = 6.878937e-05 (7.940 sec)
INFO:tensorflow:global_step/sec: 12.5607
INFO:tensorflow:loss = 1.2419947, step = 175260 (7.962 sec)
INFO:tensorflow:lr = 6.872027e-05 (7.962 sec)
INFO:tensorflow:global_step/sec: 12.602
INFO:tensorflow:loss = 1.2409464, step = 175360 (7.929 sec)
INFO:tensorflow:lr = 6.8651236e-05 (7.929 sec)
INFO:tensorflow:global_step/sec: 12.6779
INFO:tensorflow:loss = 1.2466931, step = 175460 (7.888 sec)
INFO:tensorflow:lr = 6.858228e-05 (7.888 sec)
INFO:tensorflow:global_step/sec: 12.8061
INFO:tensorflow:loss = 1.2410626, step = 175560 (7.814 sec)
INFO:tensorflow:lr = 6.8513385e-05 (7.814 sec)
INFO:tensorflow:global_step/sec: 12.7795
INFO:tensorflow:loss = 1.24094, step = 175660 (7.821 sec)
INFO:tensorflow:lr = 6.844457e-05 (7.822 sec)
INFO:tensorflow:global_step/sec: 12.6866
INFO:tensorflow:loss = 1.2434784, step = 175760 (7.881 sec)
INFO:tensorflow:lr = 6.8375804e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.7588
INFO:tensorflow:loss = 1.2445353, step = 175860 (7.841 sec)
INFO:tensorflow:lr = 6.830711e-05 (7.843 sec)
INFO:tensorflow:global_step/sec: 12.4546
INFO:tensorflow:loss = 1.2431968, step = 175960 (8.026 sec)
INFO:tensorflow:lr = 6.823851e-05 (8.025 sec)
INFO:tensorflow:Saving checkpoints for 176040 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2408674.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-176040
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-176040
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.728
INFO:tensorflow:Best (Exact Match) Accuracy: 0.731
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-176040
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 176040 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2436473, step = 176040
INFO:tensorflow:lr = 6.818367e-05
INFO:tensorflow:global_step/sec: 10.061
INFO:tensorflow:loss = 1.2422326, step = 176140 (9.946 sec)
INFO:tensorflow:lr = 6.811516e-05 (9.948 sec)
INFO:tensorflow:global_step/sec: 12.8098
INFO:tensorflow:loss = 1.2441524, step = 176240 (7.801 sec)
INFO:tensorflow:lr = 6.804675e-05 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.5074
INFO:tensorflow:loss = 1.2437139, step = 176340 (8.002 sec)
INFO:tensorflow:lr = 6.797839e-05 (8.002 sec)
INFO:tensorflow:global_step/sec: 12.7508
INFO:tensorflow:loss = 1.2430073, step = 176440 (7.841 sec)
INFO:tensorflow:lr = 6.791011e-05 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.9222
INFO:tensorflow:loss = 1.2443118, step = 176540 (7.736 sec)
INFO:tensorflow:lr = 6.784189e-05 (7.738 sec)
INFO:tensorflow:global_step/sec: 12.4978
INFO:tensorflow:loss = 1.2411668, step = 176640 (8.004 sec)
INFO:tensorflow:lr = 6.777373e-05 (8.002 sec)
INFO:tensorflow:global_step/sec: 12.7154
INFO:tensorflow:loss = 1.2438464, step = 176740 (7.867 sec)
INFO:tensorflow:lr = 6.770566e-05 (7.867 sec)
INFO:tensorflow:global_step/sec: 12.8534
INFO:tensorflow:loss = 1.2525758, step = 176840 (7.778 sec)
INFO:tensorflow:lr = 6.763765e-05 (7.778 sec)
INFO:tensorflow:global_step/sec: 12.7331
INFO:tensorflow:loss = 1.2458075, step = 176940 (7.850 sec)
INFO:tensorflow:lr = 6.756971e-05 (7.852 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.35251
INFO:tensorflow:loss = 1.2425443, step = 177040 (11.971 sec)
INFO:tensorflow:lr = 6.750183e-05 (11.970 sec)
INFO:tensorflow:global_step/sec: 12.608
INFO:tensorflow:loss = 1.2432464, step = 177140 (7.932 sec)
INFO:tensorflow:lr = 6.743402e-05 (7.932 sec)
INFO:tensorflow:global_step/sec: 12.7623
INFO:tensorflow:loss = 1.2433742, step = 177240 (7.836 sec)
INFO:tensorflow:lr = 6.736629e-05 (7.836 sec)
INFO:tensorflow:global_step/sec: 12.5755
INFO:tensorflow:loss = 1.2442988, step = 177340 (7.950 sec)
INFO:tensorflow:lr = 6.7298606e-05 (7.949 sec)
INFO:tensorflow:global_step/sec: 13.0362
INFO:tensorflow:loss = 1.2423884, step = 177440 (7.671 sec)
INFO:tensorflow:lr = 6.723101e-05 (7.672 sec)
INFO:tensorflow:global_step/sec: 12.882
INFO:tensorflow:loss = 1.2411859, step = 177540 (7.767 sec)
INFO:tensorflow:lr = 6.716346e-05 (7.767 sec)
INFO:tensorflow:global_step/sec: 12.7075
INFO:tensorflow:loss = 1.2438511, step = 177640 (7.866 sec)
INFO:tensorflow:lr = 6.7096e-05 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.7873
INFO:tensorflow:loss = 1.2472678, step = 177740 (7.824 sec)
INFO:tensorflow:lr = 6.70286e-05 (7.824 sec)
INFO:tensorflow:global_step/sec: 12.7072
INFO:tensorflow:loss = 1.2420048, step = 177840 (7.867 sec)
INFO:tensorflow:lr = 6.696127e-05 (7.866 sec)
INFO:tensorflow:global_step/sec: 12.8534
INFO:tensorflow:loss = 1.2439637, step = 177940 (7.783 sec)
INFO:tensorflow:lr = 6.6894005e-05 (7.783 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.30743
INFO:tensorflow:loss = 1.2429259, step = 178040 (12.035 sec)
INFO:tensorflow:lr = 6.6826804e-05 (12.035 sec)
INFO:tensorflow:global_step/sec: 12.8115
INFO:tensorflow:loss = 1.2417121, step = 178140 (7.803 sec)
INFO:tensorflow:lr = 6.675968e-05 (7.807 sec)
INFO:tensorflow:global_step/sec: 12.819
INFO:tensorflow:loss = 1.2494624, step = 178240 (7.801 sec)
INFO:tensorflow:lr = 6.669262e-05 (7.799 sec)
INFO:tensorflow:global_step/sec: 13.0614
INFO:tensorflow:loss = 1.2493565, step = 178340 (7.655 sec)
INFO:tensorflow:lr = 6.662562e-05 (7.654 sec)
INFO:tensorflow:global_step/sec: 12.6056
INFO:tensorflow:loss = 1.2468174, step = 178440 (7.934 sec)
INFO:tensorflow:lr = 6.65587e-05 (7.937 sec)
INFO:tensorflow:global_step/sec: 12.802
INFO:tensorflow:loss = 1.2427741, step = 178540 (7.810 sec)
INFO:tensorflow:lr = 6.649183e-05 (7.807 sec)
INFO:tensorflow:global_step/sec: 12.5317
INFO:tensorflow:loss = 1.2447419, step = 178640 (7.982 sec)
INFO:tensorflow:lr = 6.642503e-05 (7.981 sec)
INFO:tensorflow:global_step/sec: 12.833
INFO:tensorflow:loss = 1.242131, step = 178740 (7.790 sec)
INFO:tensorflow:lr = 6.635832e-05 (7.790 sec)
INFO:tensorflow:global_step/sec: 12.7559
INFO:tensorflow:loss = 1.2427136, step = 178840 (7.839 sec)
INFO:tensorflow:lr = 6.629166e-05 (7.839 sec)
INFO:tensorflow:global_step/sec: 12.829
INFO:tensorflow:loss = 1.2427148, step = 178940 (7.800 sec)
INFO:tensorflow:lr = 6.6225075e-05 (7.800 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.35557
INFO:tensorflow:loss = 1.2481025, step = 179040 (11.968 sec)
INFO:tensorflow:lr = 6.615854e-05 (11.968 sec)
INFO:tensorflow:global_step/sec: 13.0739
INFO:tensorflow:loss = 1.2435266, step = 179140 (7.647 sec)
INFO:tensorflow:lr = 6.6092085e-05 (7.646 sec)
INFO:tensorflow:global_step/sec: 12.682
INFO:tensorflow:loss = 1.2475928, step = 179240 (7.882 sec)
INFO:tensorflow:lr = 6.60257e-05 (7.883 sec)
INFO:tensorflow:global_step/sec: 12.5173
INFO:tensorflow:loss = 1.241785, step = 179340 (7.994 sec)
INFO:tensorflow:lr = 6.5959364e-05 (7.993 sec)
INFO:tensorflow:global_step/sec: 12.6721
INFO:tensorflow:loss = 1.241673, step = 179440 (7.892 sec)
INFO:tensorflow:lr = 6.589311e-05 (7.892 sec)
INFO:tensorflow:global_step/sec: 12.9129
INFO:tensorflow:loss = 1.2435246, step = 179540 (7.745 sec)
INFO:tensorflow:lr = 6.582692e-05 (7.745 sec)
INFO:tensorflow:global_step/sec: 12.656
INFO:tensorflow:loss = 1.2441323, step = 179640 (7.900 sec)
INFO:tensorflow:lr = 6.576078e-05 (7.901 sec)
INFO:tensorflow:global_step/sec: 12.9244
INFO:tensorflow:loss = 1.2419387, step = 179740 (7.732 sec)
INFO:tensorflow:lr = 6.569474e-05 (7.735 sec)
INFO:tensorflow:global_step/sec: 12.892
INFO:tensorflow:loss = 1.2468823, step = 179840 (7.758 sec)
INFO:tensorflow:lr = 6.5628745e-05 (7.756 sec)
INFO:tensorflow:global_step/sec: 12.6811
INFO:tensorflow:loss = 1.2426761, step = 179940 (7.884 sec)
INFO:tensorflow:lr = 6.556282e-05 (7.885 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.52812
INFO:tensorflow:loss = 1.2428222, step = 180040 (11.726 sec)
INFO:tensorflow:lr = 6.5496955e-05 (11.725 sec)
INFO:tensorflow:global_step/sec: 12.6168
INFO:tensorflow:loss = 1.2429457, step = 180140 (7.926 sec)
INFO:tensorflow:lr = 6.543116e-05 (7.928 sec)
INFO:tensorflow:global_step/sec: 12.7491
INFO:tensorflow:loss = 1.2432375, step = 180240 (7.847 sec)
INFO:tensorflow:lr = 6.5365435e-05 (7.847 sec)
INFO:tensorflow:global_step/sec: 12.6939
INFO:tensorflow:loss = 1.242373, step = 180340 (7.881 sec)
INFO:tensorflow:lr = 6.529978e-05 (7.880 sec)
INFO:tensorflow:global_step/sec: 12.804
INFO:tensorflow:loss = 1.2414826, step = 180440 (7.805 sec)
INFO:tensorflow:lr = 6.523419e-05 (7.807 sec)
INFO:tensorflow:global_step/sec: 12.7029
INFO:tensorflow:loss = 1.244215, step = 180540 (7.875 sec)
INFO:tensorflow:lr = 6.516865e-05 (7.874 sec)
INFO:tensorflow:global_step/sec: 12.8393
INFO:tensorflow:loss = 1.2461091, step = 180640 (7.786 sec)
INFO:tensorflow:lr = 6.510318e-05 (7.788 sec)
INFO:tensorflow:global_step/sec: 12.7857
INFO:tensorflow:loss = 1.2421577, step = 180740 (7.822 sec)
INFO:tensorflow:lr = 6.503779e-05 (7.822 sec)
INFO:tensorflow:global_step/sec: 13.0485
INFO:tensorflow:loss = 1.2420781, step = 180840 (7.668 sec)
INFO:tensorflow:lr = 6.497245e-05 (7.666 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.7315
INFO:tensorflow:loss = 1.2411906, step = 180940 (11.447 sec)
INFO:tensorflow:lr = 6.490719e-05 (11.447 sec)
INFO:tensorflow:global_step/sec: 12.7866
INFO:tensorflow:loss = 1.2427506, step = 181040 (7.825 sec)
INFO:tensorflow:lr = 6.484198e-05 (7.825 sec)
INFO:tensorflow:global_step/sec: 12.4139
INFO:tensorflow:loss = 1.2419596, step = 181140 (8.051 sec)
INFO:tensorflow:lr = 6.477685e-05 (8.051 sec)
INFO:tensorflow:global_step/sec: 12.7857
INFO:tensorflow:loss = 1.2428209, step = 181240 (7.821 sec)
INFO:tensorflow:lr = 6.4711785e-05 (7.822 sec)
INFO:tensorflow:global_step/sec: 12.76
INFO:tensorflow:loss = 1.242671, step = 181340 (7.837 sec)
INFO:tensorflow:lr = 6.464678e-05 (7.835 sec)
INFO:tensorflow:global_step/sec: 13.0807
INFO:tensorflow:loss = 1.2421178, step = 181440 (7.651 sec)
INFO:tensorflow:lr = 6.458184e-05 (7.651 sec)
INFO:tensorflow:global_step/sec: 12.8754
INFO:tensorflow:loss = 1.2421011, step = 181540 (7.763 sec)
INFO:tensorflow:lr = 6.4516964e-05 (7.764 sec)
INFO:tensorflow:global_step/sec: 12.3776
INFO:tensorflow:loss = 1.2435634, step = 181640 (8.076 sec)
INFO:tensorflow:lr = 6.445216e-05 (8.076 sec)
INFO:tensorflow:global_step/sec: 12.6016
INFO:tensorflow:loss = 1.2422757, step = 181740 (7.936 sec)
INFO:tensorflow:lr = 6.4387415e-05 (7.940 sec)
INFO:tensorflow:global_step/sec: 12.6405
INFO:tensorflow:loss = 1.2412153, step = 181840 (7.915 sec)
INFO:tensorflow:lr = 6.432273e-05 (7.911 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.44978
INFO:tensorflow:loss = 1.244641, step = 181940 (10.582 sec)
INFO:tensorflow:lr = 6.425812e-05 (10.582 sec)
INFO:tensorflow:global_step/sec: 13.0937
INFO:tensorflow:loss = 1.2411733, step = 182040 (7.638 sec)
INFO:tensorflow:lr = 6.419357e-05 (7.638 sec)
INFO:tensorflow:global_step/sec: 12.9868
INFO:tensorflow:loss = 1.2436959, step = 182140 (7.700 sec)
INFO:tensorflow:lr = 6.412908e-05 (7.700 sec)
INFO:tensorflow:global_step/sec: 12.7698
INFO:tensorflow:loss = 1.2463115, step = 182240 (7.826 sec)
INFO:tensorflow:lr = 6.406467e-05 (7.830 sec)
INFO:tensorflow:global_step/sec: 12.803
INFO:tensorflow:loss = 1.2440659, step = 182340 (7.816 sec)
INFO:tensorflow:lr = 6.4000305e-05 (7.812 sec)
INFO:tensorflow:global_step/sec: 12.522
INFO:tensorflow:loss = 1.2420621, step = 182440 (7.985 sec)
INFO:tensorflow:lr = 6.393602e-05 (7.985 sec)
INFO:tensorflow:global_step/sec: 12.4941
INFO:tensorflow:loss = 1.2414142, step = 182540 (8.005 sec)
INFO:tensorflow:lr = 6.387179e-05 (8.005 sec)
INFO:tensorflow:global_step/sec: 12.6789
INFO:tensorflow:loss = 1.2461977, step = 182640 (7.885 sec)
INFO:tensorflow:lr = 6.380763e-05 (7.886 sec)
INFO:tensorflow:global_step/sec: 13.0701
INFO:tensorflow:loss = 1.2416025, step = 182740 (7.654 sec)
INFO:tensorflow:lr = 6.3743544e-05 (7.652 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.29
INFO:tensorflow:loss = 1.2431751, step = 182840 (8.133 sec)
INFO:tensorflow:lr = 6.36795e-05 (8.136 sec)
INFO:tensorflow:global_step/sec: 10.6503
INFO:tensorflow:loss = 1.2422962, step = 182940 (9.393 sec)
INFO:tensorflow:lr = 6.361554e-05 (9.390 sec)
INFO:tensorflow:global_step/sec: 13.0235
INFO:tensorflow:loss = 1.2496976, step = 183040 (7.673 sec)
INFO:tensorflow:lr = 6.355163e-05 (7.677 sec)
INFO:tensorflow:global_step/sec: 12.6265
INFO:tensorflow:loss = 1.2432393, step = 183140 (7.924 sec)
INFO:tensorflow:lr = 6.348779e-05 (7.920 sec)
INFO:tensorflow:global_step/sec: 12.6644
INFO:tensorflow:loss = 1.243746, step = 183240 (7.891 sec)
INFO:tensorflow:lr = 6.342402e-05 (7.892 sec)
INFO:tensorflow:global_step/sec: 12.7925
INFO:tensorflow:loss = 1.2410957, step = 183340 (7.817 sec)
INFO:tensorflow:lr = 6.3360305e-05 (7.818 sec)
INFO:tensorflow:global_step/sec: 13.0226
INFO:tensorflow:loss = 1.2438482, step = 183440 (7.684 sec)
INFO:tensorflow:lr = 6.329667e-05 (7.684 sec)
INFO:tensorflow:global_step/sec: 12.7638
INFO:tensorflow:loss = 1.245033, step = 183540 (7.835 sec)
INFO:tensorflow:lr = 6.3233085e-05 (7.835 sec)
INFO:tensorflow:global_step/sec: 12.7621
INFO:tensorflow:loss = 1.2443069, step = 183640 (7.837 sec)
INFO:tensorflow:lr = 6.316956e-05 (7.837 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.5158
INFO:tensorflow:loss = 1.2428632, step = 183740 (7.991 sec)
INFO:tensorflow:lr = 6.31061e-05 (7.990 sec)
INFO:tensorflow:global_step/sec: 11.0577
INFO:tensorflow:loss = 1.2430586, step = 183840 (9.036 sec)
INFO:tensorflow:lr = 6.304271e-05 (9.036 sec)
INFO:tensorflow:global_step/sec: 12.7589
INFO:tensorflow:loss = 1.242105, step = 183940 (7.842 sec)
INFO:tensorflow:lr = 6.297939e-05 (7.842 sec)
INFO:tensorflow:global_step/sec: 12.875
INFO:tensorflow:loss = 1.2421196, step = 184040 (7.770 sec)
INFO:tensorflow:lr = 6.2916115e-05 (7.770 sec)
INFO:tensorflow:global_step/sec: 12.8843
INFO:tensorflow:loss = 1.2427078, step = 184140 (7.759 sec)
INFO:tensorflow:lr = 6.285291e-05 (7.758 sec)
INFO:tensorflow:global_step/sec: 12.8214
INFO:tensorflow:loss = 1.2439513, step = 184240 (7.795 sec)
INFO:tensorflow:lr = 6.278978e-05 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.6415
INFO:tensorflow:loss = 1.2408562, step = 184340 (7.911 sec)
INFO:tensorflow:lr = 6.272671e-05 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.6495
INFO:tensorflow:loss = 1.2422056, step = 184440 (7.915 sec)
INFO:tensorflow:lr = 6.2663705e-05 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.7647
INFO:tensorflow:loss = 1.2462195, step = 184540 (7.828 sec)
INFO:tensorflow:lr = 6.260074e-05 (7.828 sec)
INFO:tensorflow:global_step/sec: 12.8077
INFO:tensorflow:loss = 1.2462376, step = 184640 (7.804 sec)
INFO:tensorflow:lr = 6.253786e-05 (7.804 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1954
INFO:tensorflow:loss = 1.2416941, step = 184740 (8.206 sec)
INFO:tensorflow:lr = 6.247504e-05 (8.209 sec)
INFO:tensorflow:global_step/sec: 11.4304
INFO:tensorflow:loss = 1.243738, step = 184840 (8.742 sec)
INFO:tensorflow:lr = 6.2412284e-05 (8.739 sec)
INFO:tensorflow:global_step/sec: 12.7942
INFO:tensorflow:loss = 1.242184, step = 184940 (7.818 sec)
INFO:tensorflow:lr = 6.23496e-05 (7.818 sec)
INFO:tensorflow:global_step/sec: 12.5375
INFO:tensorflow:loss = 1.2413771, step = 185040 (7.977 sec)
INFO:tensorflow:lr = 6.2286956e-05 (7.977 sec)
INFO:tensorflow:global_step/sec: 12.7844
INFO:tensorflow:loss = 1.2429293, step = 185140 (7.819 sec)
INFO:tensorflow:lr = 6.222439e-05 (7.819 sec)
INFO:tensorflow:global_step/sec: 12.7958
INFO:tensorflow:loss = 1.245569, step = 185240 (7.820 sec)
INFO:tensorflow:lr = 6.216188e-05 (7.820 sec)
INFO:tensorflow:global_step/sec: 12.5349
INFO:tensorflow:loss = 1.240598, step = 185340 (7.979 sec)
INFO:tensorflow:lr = 6.209943e-05 (7.979 sec)
INFO:tensorflow:global_step/sec: 12.7825
INFO:tensorflow:loss = 1.2407473, step = 185440 (7.820 sec)
INFO:tensorflow:lr = 6.203707e-05 (7.821 sec)
INFO:tensorflow:global_step/sec: 12.7777
INFO:tensorflow:loss = 1.24277, step = 185540 (7.826 sec)
INFO:tensorflow:lr = 6.1974744e-05 (7.825 sec)
INFO:tensorflow:global_step/sec: 12.9954
INFO:tensorflow:loss = 1.2449442, step = 185640 (7.693 sec)
INFO:tensorflow:lr = 6.1912484e-05 (7.693 sec)
INFO:tensorflow:global_step/sec: 12.6443
INFO:tensorflow:loss = 1.2449242, step = 185740 (7.907 sec)
INFO:tensorflow:lr = 6.1850296e-05 (7.907 sec)
INFO:tensorflow:Saving checkpoints for 185820 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.242429.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-185820
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-185820
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.732
INFO:tensorflow:Best (Exact Match) Accuracy: 0.732
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-185820
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 185820 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2430956, step = 185820
INFO:tensorflow:lr = 6.180058e-05
INFO:tensorflow:global_step/sec: 10.2397
INFO:tensorflow:loss = 1.2422625, step = 185920 (9.768 sec)
INFO:tensorflow:lr = 6.17385e-05 (9.768 sec)
INFO:tensorflow:global_step/sec: 12.5678
INFO:tensorflow:loss = 1.2413614, step = 186020 (7.957 sec)
INFO:tensorflow:lr = 6.167648e-05 (7.956 sec)
INFO:tensorflow:global_step/sec: 12.6587
INFO:tensorflow:loss = 1.242497, step = 186120 (7.899 sec)
INFO:tensorflow:lr = 6.1614526e-05 (7.899 sec)
INFO:tensorflow:global_step/sec: 13.0192
INFO:tensorflow:loss = 1.2470552, step = 186220 (7.683 sec)
INFO:tensorflow:lr = 6.155264e-05 (7.683 sec)
INFO:tensorflow:global_step/sec: 12.8884
INFO:tensorflow:loss = 1.2431347, step = 186320 (7.758 sec)
INFO:tensorflow:lr = 6.1490806e-05 (7.758 sec)
INFO:tensorflow:global_step/sec: 12.7696
INFO:tensorflow:loss = 1.2459142, step = 186420 (7.830 sec)
INFO:tensorflow:lr = 6.142903e-05 (7.832 sec)
INFO:tensorflow:global_step/sec: 12.9001
INFO:tensorflow:loss = 1.2413319, step = 186520 (7.754 sec)
INFO:tensorflow:lr = 6.136733e-05 (7.755 sec)
INFO:tensorflow:global_step/sec: 12.7284
INFO:tensorflow:loss = 1.2430797, step = 186620 (7.859 sec)
INFO:tensorflow:lr = 6.130568e-05 (7.856 sec)
INFO:tensorflow:global_step/sec: 12.3593
INFO:tensorflow:loss = 1.2427655, step = 186720 (8.087 sec)
INFO:tensorflow:lr = 6.12441e-05 (8.087 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.38791
INFO:tensorflow:loss = 1.2434381, step = 186820 (11.926 sec)
INFO:tensorflow:lr = 6.118258e-05 (11.926 sec)
INFO:tensorflow:global_step/sec: 12.7746
INFO:tensorflow:loss = 1.2431988, step = 186920 (7.828 sec)
INFO:tensorflow:lr = 6.1121114e-05 (7.828 sec)
INFO:tensorflow:global_step/sec: 12.3576
INFO:tensorflow:loss = 1.240415, step = 187020 (8.090 sec)
INFO:tensorflow:lr = 6.105973e-05 (8.092 sec)
INFO:tensorflow:global_step/sec: 12.5951
INFO:tensorflow:loss = 1.2401947, step = 187120 (7.938 sec)
INFO:tensorflow:lr = 6.0998384e-05 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.5726
INFO:tensorflow:loss = 1.2473856, step = 187220 (7.957 sec)
INFO:tensorflow:lr = 6.0937105e-05 (7.953 sec)
INFO:tensorflow:global_step/sec: 12.7697
INFO:tensorflow:loss = 1.2429252, step = 187320 (7.828 sec)
INFO:tensorflow:lr = 6.0875893e-05 (7.825 sec)
INFO:tensorflow:global_step/sec: 13.0692
INFO:tensorflow:loss = 1.2432778, step = 187420 (7.652 sec)
INFO:tensorflow:lr = 6.0814742e-05 (7.652 sec)
INFO:tensorflow:global_step/sec: 12.8313
INFO:tensorflow:loss = 1.24176, step = 187520 (7.793 sec)
INFO:tensorflow:lr = 6.075366e-05 (7.797 sec)
INFO:tensorflow:global_step/sec: 12.7361
INFO:tensorflow:loss = 1.2408872, step = 187620 (7.858 sec)
INFO:tensorflow:lr = 6.0692626e-05 (7.853 sec)
INFO:tensorflow:global_step/sec: 12.5969
INFO:tensorflow:loss = 1.2419, step = 187720 (7.933 sec)
INFO:tensorflow:lr = 6.0631653e-05 (7.934 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.32018
INFO:tensorflow:loss = 1.2422671, step = 187820 (12.022 sec)
INFO:tensorflow:lr = 6.0570757e-05 (12.021 sec)
INFO:tensorflow:global_step/sec: 13.0867
INFO:tensorflow:loss = 1.2460558, step = 187920 (7.638 sec)
INFO:tensorflow:lr = 6.05099e-05 (7.638 sec)
INFO:tensorflow:global_step/sec: 12.3994
INFO:tensorflow:loss = 1.2414271, step = 188020 (8.072 sec)
INFO:tensorflow:lr = 6.044912e-05 (8.072 sec)
INFO:tensorflow:global_step/sec: 12.533
INFO:tensorflow:loss = 1.2409451, step = 188120 (7.975 sec)
INFO:tensorflow:lr = 6.0388404e-05 (7.976 sec)
INFO:tensorflow:global_step/sec: 12.8062
INFO:tensorflow:loss = 1.2424439, step = 188220 (7.806 sec)
INFO:tensorflow:lr = 6.032773e-05 (7.805 sec)
INFO:tensorflow:global_step/sec: 12.7571
INFO:tensorflow:loss = 1.2418518, step = 188320 (7.842 sec)
INFO:tensorflow:lr = 6.0267135e-05 (7.841 sec)
INFO:tensorflow:global_step/sec: 12.7492
INFO:tensorflow:loss = 1.2417756, step = 188420 (7.845 sec)
INFO:tensorflow:lr = 6.0206596e-05 (7.846 sec)
INFO:tensorflow:global_step/sec: 12.6351
INFO:tensorflow:loss = 1.2414497, step = 188520 (7.912 sec)
INFO:tensorflow:lr = 6.014612e-05 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.8196
INFO:tensorflow:loss = 1.242274, step = 188620 (7.805 sec)
INFO:tensorflow:lr = 6.0085706e-05 (7.803 sec)
INFO:tensorflow:global_step/sec: 12.6623
INFO:tensorflow:loss = 1.2413298, step = 188720 (7.895 sec)
INFO:tensorflow:lr = 6.0025337e-05 (7.895 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.37098
INFO:tensorflow:loss = 1.241566, step = 188820 (11.942 sec)
INFO:tensorflow:lr = 5.996505e-05 (11.941 sec)
INFO:tensorflow:global_step/sec: 12.5102
INFO:tensorflow:loss = 1.2479603, step = 188920 (7.998 sec)
INFO:tensorflow:lr = 5.9904807e-05 (8.003 sec)
INFO:tensorflow:global_step/sec: 12.7268
INFO:tensorflow:loss = 1.2499129, step = 189020 (7.855 sec)
INFO:tensorflow:lr = 5.9844642e-05 (7.851 sec)
INFO:tensorflow:global_step/sec: 12.705
INFO:tensorflow:loss = 1.2405927, step = 189120 (7.875 sec)
INFO:tensorflow:lr = 5.9784517e-05 (7.875 sec)
INFO:tensorflow:global_step/sec: 12.8384
INFO:tensorflow:loss = 1.2449517, step = 189220 (7.784 sec)
INFO:tensorflow:lr = 5.9724454e-05 (7.784 sec)
INFO:tensorflow:global_step/sec: 12.8968
INFO:tensorflow:loss = 1.2450413, step = 189320 (7.756 sec)
INFO:tensorflow:lr = 5.9664464e-05 (7.758 sec)
INFO:tensorflow:global_step/sec: 12.9367
INFO:tensorflow:loss = 1.2425894, step = 189420 (7.727 sec)
INFO:tensorflow:lr = 5.960453e-05 (7.726 sec)
INFO:tensorflow:global_step/sec: 12.5036
INFO:tensorflow:loss = 1.2416962, step = 189520 (8.003 sec)
INFO:tensorflow:lr = 5.954466e-05 (8.003 sec)
INFO:tensorflow:global_step/sec: 12.7556
INFO:tensorflow:loss = 1.2433455, step = 189620 (7.835 sec)
INFO:tensorflow:lr = 5.948484e-05 (7.834 sec)
INFO:tensorflow:global_step/sec: 13.0041
INFO:tensorflow:loss = 1.2446123, step = 189720 (7.690 sec)
INFO:tensorflow:lr = 5.9425085e-05 (7.690 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.51546
INFO:tensorflow:loss = 1.2419295, step = 189820 (11.748 sec)
INFO:tensorflow:lr = 5.9365397e-05 (11.747 sec)
INFO:tensorflow:global_step/sec: 12.8369
INFO:tensorflow:loss = 1.2428107, step = 189920 (7.784 sec)
INFO:tensorflow:lr = 5.9305763e-05 (7.784 sec)
INFO:tensorflow:global_step/sec: 12.8227
INFO:tensorflow:loss = 1.2412173, step = 190020 (7.799 sec)
INFO:tensorflow:lr = 5.9246187e-05 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.8477
INFO:tensorflow:loss = 1.2426255, step = 190120 (7.788 sec)
INFO:tensorflow:lr = 5.918667e-05 (7.787 sec)
INFO:tensorflow:global_step/sec: 13.0337
INFO:tensorflow:loss = 1.2452132, step = 190220 (7.670 sec)
INFO:tensorflow:lr = 5.9127215e-05 (7.672 sec)
INFO:tensorflow:global_step/sec: 12.5826
INFO:tensorflow:loss = 1.2415522, step = 190320 (7.944 sec)
INFO:tensorflow:lr = 5.9067825e-05 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.4786
INFO:tensorflow:loss = 1.2416269, step = 190420 (8.014 sec)
INFO:tensorflow:lr = 5.900849e-05 (8.015 sec)
INFO:tensorflow:global_step/sec: 12.8258
INFO:tensorflow:loss = 1.2427078, step = 190520 (7.804 sec)
INFO:tensorflow:lr = 5.894922e-05 (7.803 sec)
INFO:tensorflow:global_step/sec: 12.7805
INFO:tensorflow:loss = 1.2445235, step = 190620 (7.821 sec)
INFO:tensorflow:lr = 5.8889997e-05 (7.820 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.49193
INFO:tensorflow:loss = 1.247068, step = 190720 (11.774 sec)
INFO:tensorflow:lr = 5.883084e-05 (11.775 sec)
INFO:tensorflow:global_step/sec: 12.6129
INFO:tensorflow:loss = 1.2422463, step = 190820 (7.928 sec)
INFO:tensorflow:lr = 5.877174e-05 (7.929 sec)
INFO:tensorflow:global_step/sec: 12.6408
INFO:tensorflow:loss = 1.2405804, step = 190920 (7.911 sec)
INFO:tensorflow:lr = 5.8712703e-05 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.635
INFO:tensorflow:loss = 1.2446344, step = 191020 (7.914 sec)
INFO:tensorflow:lr = 5.8653728e-05 (7.914 sec)
INFO:tensorflow:global_step/sec: 12.713
INFO:tensorflow:loss = 1.2437266, step = 191120 (7.866 sec)
INFO:tensorflow:lr = 5.8594804e-05 (7.869 sec)
INFO:tensorflow:global_step/sec: 12.8576
INFO:tensorflow:loss = 1.2444649, step = 191220 (7.782 sec)
INFO:tensorflow:lr = 5.8535945e-05 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.7873
INFO:tensorflow:loss = 1.2414025, step = 191320 (7.821 sec)
INFO:tensorflow:lr = 5.847715e-05 (7.821 sec)
INFO:tensorflow:global_step/sec: 12.5444
INFO:tensorflow:loss = 1.2417895, step = 191420 (7.970 sec)
INFO:tensorflow:lr = 5.8418398e-05 (7.970 sec)
INFO:tensorflow:global_step/sec: 12.6289
INFO:tensorflow:loss = 1.2412646, step = 191520 (7.916 sec)
INFO:tensorflow:lr = 5.835972e-05 (7.915 sec)
INFO:tensorflow:global_step/sec: 12.7265
INFO:tensorflow:loss = 1.2413337, step = 191620 (7.859 sec)
INFO:tensorflow:lr = 5.8301095e-05 (7.861 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.09703
INFO:tensorflow:loss = 1.2453784, step = 191720 (10.996 sec)
INFO:tensorflow:lr = 5.8242527e-05 (10.994 sec)
INFO:tensorflow:global_step/sec: 12.7438
INFO:tensorflow:loss = 1.2416813, step = 191820 (7.849 sec)
INFO:tensorflow:lr = 5.8184025e-05 (7.849 sec)
INFO:tensorflow:global_step/sec: 13.0111
INFO:tensorflow:loss = 1.2428048, step = 191920 (7.680 sec)
INFO:tensorflow:lr = 5.8125574e-05 (7.682 sec)
INFO:tensorflow:global_step/sec: 12.8564
INFO:tensorflow:loss = 1.2413689, step = 192020 (7.777 sec)
INFO:tensorflow:lr = 5.8067188e-05 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.5216
INFO:tensorflow:loss = 1.2429854, step = 192120 (7.986 sec)
INFO:tensorflow:lr = 5.8008867e-05 (7.982 sec)
INFO:tensorflow:global_step/sec: 12.8034
INFO:tensorflow:loss = 1.244082, step = 192220 (7.814 sec)
INFO:tensorflow:lr = 5.7950583e-05 (7.813 sec)
INFO:tensorflow:global_step/sec: 12.5927
INFO:tensorflow:loss = 1.2440885, step = 192320 (7.944 sec)
INFO:tensorflow:lr = 5.7892375e-05 (7.947 sec)
INFO:tensorflow:global_step/sec: 12.5696
INFO:tensorflow:loss = 1.2455595, step = 192420 (7.949 sec)
INFO:tensorflow:lr = 5.783422e-05 (7.948 sec)
INFO:tensorflow:global_step/sec: 12.8255
INFO:tensorflow:loss = 1.2410458, step = 192520 (7.797 sec)
INFO:tensorflow:lr = 5.7776128e-05 (7.796 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.2228
INFO:tensorflow:loss = 1.2421664, step = 192620 (8.181 sec)
INFO:tensorflow:lr = 5.771809e-05 (8.180 sec)
INFO:tensorflow:global_step/sec: 10.7852
INFO:tensorflow:loss = 1.2432908, step = 192720 (9.272 sec)
INFO:tensorflow:lr = 5.76601e-05 (9.272 sec)
INFO:tensorflow:global_step/sec: 12.9277
INFO:tensorflow:loss = 1.2420142, step = 192820 (7.736 sec)
INFO:tensorflow:lr = 5.7602185e-05 (7.736 sec)
INFO:tensorflow:global_step/sec: 13.0745
INFO:tensorflow:loss = 1.2412543, step = 192920 (7.649 sec)
INFO:tensorflow:lr = 5.7544326e-05 (7.649 sec)
INFO:tensorflow:global_step/sec: 12.769
INFO:tensorflow:loss = 1.2427813, step = 193020 (7.836 sec)
INFO:tensorflow:lr = 5.7486523e-05 (7.835 sec)
INFO:tensorflow:global_step/sec: 12.6918
INFO:tensorflow:loss = 1.2403716, step = 193120 (7.874 sec)
INFO:tensorflow:lr = 5.7428777e-05 (7.874 sec)
INFO:tensorflow:global_step/sec: 12.6062
INFO:tensorflow:loss = 1.2418182, step = 193220 (7.933 sec)
INFO:tensorflow:lr = 5.737108e-05 (7.932 sec)
INFO:tensorflow:global_step/sec: 12.3039
INFO:tensorflow:loss = 1.2434244, step = 193320 (8.128 sec)
INFO:tensorflow:lr = 5.7313453e-05 (8.129 sec)
INFO:tensorflow:global_step/sec: 12.7618
INFO:tensorflow:loss = 1.2468263, step = 193420 (7.836 sec)
INFO:tensorflow:lr = 5.7255882e-05 (7.835 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.9098
INFO:tensorflow:loss = 1.2425442, step = 193520 (7.748 sec)
INFO:tensorflow:lr = 5.719837e-05 (7.749 sec)
INFO:tensorflow:global_step/sec: 11.04
INFO:tensorflow:loss = 1.2404422, step = 193620 (9.060 sec)
INFO:tensorflow:lr = 5.7140904e-05 (9.059 sec)
INFO:tensorflow:global_step/sec: 12.8864
INFO:tensorflow:loss = 1.2420374, step = 193720 (7.763 sec)
INFO:tensorflow:lr = 5.7083504e-05 (7.763 sec)
INFO:tensorflow:global_step/sec: 12.6563
INFO:tensorflow:loss = 1.24243, step = 193820 (7.902 sec)
INFO:tensorflow:lr = 5.7026165e-05 (7.901 sec)
INFO:tensorflow:global_step/sec: 12.5789
INFO:tensorflow:loss = 1.2422926, step = 193920 (7.943 sec)
INFO:tensorflow:lr = 5.696888e-05 (7.944 sec)
INFO:tensorflow:global_step/sec: 12.7081
INFO:tensorflow:loss = 1.2414744, step = 194020 (7.869 sec)
INFO:tensorflow:lr = 5.6911656e-05 (7.868 sec)
INFO:tensorflow:global_step/sec: 12.9398
INFO:tensorflow:loss = 1.242649, step = 194120 (7.728 sec)
INFO:tensorflow:lr = 5.6854486e-05 (7.727 sec)
INFO:tensorflow:global_step/sec: 13.0157
INFO:tensorflow:loss = 1.2424544, step = 194220 (7.691 sec)
INFO:tensorflow:lr = 5.6797366e-05 (7.692 sec)
INFO:tensorflow:global_step/sec: 12.5769
INFO:tensorflow:loss = 1.2423269, step = 194320 (7.943 sec)
INFO:tensorflow:lr = 5.674032e-05 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.6673
INFO:tensorflow:loss = 1.242586, step = 194420 (7.902 sec)
INFO:tensorflow:lr = 5.668332e-05 (7.902 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.923
INFO:tensorflow:loss = 1.2406418, step = 194520 (8.380 sec)
INFO:tensorflow:lr = 5.6626384e-05 (8.380 sec)
INFO:tensorflow:global_step/sec: 11.3406
INFO:tensorflow:loss = 1.2411753, step = 194620 (8.823 sec)
INFO:tensorflow:lr = 5.656949e-05 (8.823 sec)
INFO:tensorflow:global_step/sec: 12.7301
INFO:tensorflow:loss = 1.2445536, step = 194720 (7.855 sec)
INFO:tensorflow:lr = 5.6512672e-05 (7.859 sec)
INFO:tensorflow:global_step/sec: 12.9083
INFO:tensorflow:loss = 1.241786, step = 194820 (7.748 sec)
INFO:tensorflow:lr = 5.64559e-05 (7.743 sec)
INFO:tensorflow:global_step/sec: 12.7429
INFO:tensorflow:loss = 1.2426987, step = 194920 (7.845 sec)
INFO:tensorflow:lr = 5.639918e-05 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.804
INFO:tensorflow:loss = 1.2429112, step = 195020 (7.815 sec)
INFO:tensorflow:lr = 5.6342542e-05 (7.812 sec)
INFO:tensorflow:global_step/sec: 12.4789
INFO:tensorflow:loss = 1.2444768, step = 195120 (8.011 sec)
INFO:tensorflow:lr = 5.628594e-05 (8.012 sec)
INFO:tensorflow:global_step/sec: 12.7206
INFO:tensorflow:loss = 1.2421489, step = 195220 (7.856 sec)
INFO:tensorflow:lr = 5.62294e-05 (7.857 sec)
INFO:tensorflow:global_step/sec: 12.4673
INFO:tensorflow:loss = 1.2470261, step = 195320 (8.025 sec)
INFO:tensorflow:lr = 5.6172918e-05 (8.024 sec)
INFO:tensorflow:global_step/sec: 12.5697
INFO:tensorflow:loss = 1.243064, step = 195420 (7.954 sec)
INFO:tensorflow:lr = 5.6116485e-05 (7.954 sec)
INFO:tensorflow:global_step/sec: 12.7259
INFO:tensorflow:loss = 1.2425088, step = 195520 (7.854 sec)
INFO:tensorflow:lr = 5.6060122e-05 (7.859 sec)
INFO:tensorflow:Saving checkpoints for 195600 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2410872.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-195600
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-195600
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.733
INFO:tensorflow:Best (Exact Match) Accuracy: 0.733
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-195600
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 195600 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2424295, step = 195600
INFO:tensorflow:lr = 5.6015062e-05
INFO:tensorflow:global_step/sec: 10.1273
INFO:tensorflow:loss = 1.242512, step = 195700 (9.884 sec)
INFO:tensorflow:lr = 5.5958793e-05 (9.884 sec)
INFO:tensorflow:global_step/sec: 12.6818
INFO:tensorflow:loss = 1.2421359, step = 195800 (7.884 sec)
INFO:tensorflow:lr = 5.5902587e-05 (7.885 sec)
INFO:tensorflow:global_step/sec: 12.732
INFO:tensorflow:loss = 1.2427982, step = 195900 (7.855 sec)
INFO:tensorflow:lr = 5.584642e-05 (7.855 sec)
INFO:tensorflow:global_step/sec: 12.7171
INFO:tensorflow:loss = 1.2436755, step = 196000 (7.861 sec)
INFO:tensorflow:lr = 5.5790315e-05 (7.860 sec)
INFO:tensorflow:global_step/sec: 12.8571
INFO:tensorflow:loss = 1.242795, step = 196100 (7.779 sec)
INFO:tensorflow:lr = 5.5734283e-05 (7.778 sec)
INFO:tensorflow:global_step/sec: 12.7439
INFO:tensorflow:loss = 1.2420249, step = 196200 (7.846 sec)
INFO:tensorflow:lr = 5.5678294e-05 (7.846 sec)
INFO:tensorflow:global_step/sec: 12.7956
INFO:tensorflow:loss = 1.2426386, step = 196300 (7.813 sec)
INFO:tensorflow:lr = 5.5622368e-05 (7.812 sec)
INFO:tensorflow:global_step/sec: 12.7607
INFO:tensorflow:loss = 1.2421803, step = 196400 (7.835 sec)
INFO:tensorflow:lr = 5.556649e-05 (7.839 sec)
INFO:tensorflow:global_step/sec: 12.9493
INFO:tensorflow:loss = 1.2418551, step = 196500 (7.729 sec)
INFO:tensorflow:lr = 5.5510667e-05 (7.725 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.32006
INFO:tensorflow:loss = 1.2419361, step = 196600 (12.016 sec)
INFO:tensorflow:lr = 5.5454908e-05 (12.017 sec)
INFO:tensorflow:global_step/sec: 12.5265
INFO:tensorflow:loss = 1.2419522, step = 196700 (7.985 sec)
INFO:tensorflow:lr = 5.5399207e-05 (7.985 sec)
INFO:tensorflow:global_step/sec: 12.6023
INFO:tensorflow:loss = 1.2418892, step = 196800 (7.930 sec)
INFO:tensorflow:lr = 5.534356e-05 (7.929 sec)
INFO:tensorflow:global_step/sec: 12.79
INFO:tensorflow:loss = 1.2402346, step = 196900 (7.818 sec)
INFO:tensorflow:lr = 5.5287957e-05 (7.818 sec)
INFO:tensorflow:global_step/sec: 12.8103
INFO:tensorflow:loss = 1.242253, step = 197000 (7.807 sec)
INFO:tensorflow:lr = 5.5232424e-05 (7.807 sec)
INFO:tensorflow:global_step/sec: 12.9615
INFO:tensorflow:loss = 1.2404039, step = 197100 (7.719 sec)
INFO:tensorflow:lr = 5.517694e-05 (7.718 sec)
INFO:tensorflow:global_step/sec: 12.904
INFO:tensorflow:loss = 1.2524287, step = 197200 (7.751 sec)
INFO:tensorflow:lr = 5.5121516e-05 (7.751 sec)
INFO:tensorflow:global_step/sec: 12.6763
INFO:tensorflow:loss = 1.2418357, step = 197300 (7.887 sec)
INFO:tensorflow:lr = 5.506615e-05 (7.887 sec)
INFO:tensorflow:global_step/sec: 12.6131
INFO:tensorflow:loss = 1.2412477, step = 197400 (7.925 sec)
INFO:tensorflow:lr = 5.5010823e-05 (7.924 sec)
INFO:tensorflow:global_step/sec: 12.6833
INFO:tensorflow:loss = 1.2424623, step = 197500 (7.889 sec)
INFO:tensorflow:lr = 5.495556e-05 (7.890 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.2987
INFO:tensorflow:loss = 1.2423265, step = 197600 (12.045 sec)
INFO:tensorflow:lr = 5.4900363e-05 (12.046 sec)
INFO:tensorflow:global_step/sec: 13.031
INFO:tensorflow:loss = 1.2457525, step = 197700 (7.678 sec)
INFO:tensorflow:lr = 5.484521e-05 (7.679 sec)
INFO:tensorflow:global_step/sec: 12.9439
INFO:tensorflow:loss = 1.2480465, step = 197800 (7.721 sec)
INFO:tensorflow:lr = 5.4790125e-05 (7.720 sec)
INFO:tensorflow:global_step/sec: 12.7214
INFO:tensorflow:loss = 1.2456297, step = 197900 (7.862 sec)
INFO:tensorflow:lr = 5.4735083e-05 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.8302
INFO:tensorflow:loss = 1.2427973, step = 198000 (7.799 sec)
INFO:tensorflow:lr = 5.4680095e-05 (7.799 sec)
INFO:tensorflow:global_step/sec: 12.8085
INFO:tensorflow:loss = 1.2444547, step = 198100 (7.801 sec)
INFO:tensorflow:lr = 5.4625176e-05 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.8353
INFO:tensorflow:loss = 1.2432563, step = 198200 (7.796 sec)
INFO:tensorflow:lr = 5.4570293e-05 (7.798 sec)
INFO:tensorflow:global_step/sec: 12.6867
INFO:tensorflow:loss = 1.2415416, step = 198300 (7.877 sec)
INFO:tensorflow:lr = 5.451548e-05 (7.878 sec)
INFO:tensorflow:global_step/sec: 12.7024
INFO:tensorflow:loss = 1.2426435, step = 198400 (7.873 sec)
INFO:tensorflow:lr = 5.446072e-05 (7.870 sec)
INFO:tensorflow:global_step/sec: 12.6592
INFO:tensorflow:loss = 1.2404841, step = 198500 (7.905 sec)
INFO:tensorflow:lr = 5.4406017e-05 (7.905 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.28541
INFO:tensorflow:loss = 1.2406154, step = 198600 (12.064 sec)
INFO:tensorflow:lr = 5.4351363e-05 (12.064 sec)
INFO:tensorflow:global_step/sec: 13.0017
INFO:tensorflow:loss = 1.2456881, step = 198700 (7.697 sec)
INFO:tensorflow:lr = 5.429676e-05 (7.697 sec)
INFO:tensorflow:global_step/sec: 12.81
INFO:tensorflow:loss = 1.2410623, step = 198800 (7.805 sec)
INFO:tensorflow:lr = 5.424222e-05 (7.806 sec)
INFO:tensorflow:global_step/sec: 12.8692
INFO:tensorflow:loss = 1.2445118, step = 198900 (7.766 sec)
INFO:tensorflow:lr = 5.4187738e-05 (7.766 sec)
INFO:tensorflow:global_step/sec: 12.7127
INFO:tensorflow:loss = 1.244541, step = 199000 (7.871 sec)
INFO:tensorflow:lr = 5.41333e-05 (7.870 sec)
INFO:tensorflow:global_step/sec: 12.7484
INFO:tensorflow:loss = 1.2432263, step = 199100 (7.845 sec)
INFO:tensorflow:lr = 5.407892e-05 (7.844 sec)
INFO:tensorflow:global_step/sec: 12.6662
INFO:tensorflow:loss = 1.2418202, step = 199200 (7.895 sec)
INFO:tensorflow:lr = 5.4024593e-05 (7.895 sec)
INFO:tensorflow:global_step/sec: 13.0052
INFO:tensorflow:loss = 1.2434285, step = 199300 (7.686 sec)
INFO:tensorflow:lr = 5.397033e-05 (7.687 sec)
INFO:tensorflow:global_step/sec: 12.6949
INFO:tensorflow:loss = 1.241352, step = 199400 (7.881 sec)
INFO:tensorflow:lr = 5.3916112e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.5404
INFO:tensorflow:loss = 1.2433487, step = 199500 (7.969 sec)
INFO:tensorflow:lr = 5.3861946e-05 (7.976 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.31808
INFO:tensorflow:loss = 1.2508881, step = 199600 (12.024 sec)
INFO:tensorflow:lr = 5.3807846e-05 (12.018 sec)
INFO:tensorflow:global_step/sec: 13.2693
INFO:tensorflow:loss = 1.2405204, step = 199700 (7.538 sec)
INFO:tensorflow:lr = 5.3753793e-05 (7.536 sec)
INFO:tensorflow:global_step/sec: 12.5754
INFO:tensorflow:loss = 1.2424177, step = 199800 (7.951 sec)
INFO:tensorflow:lr = 5.36998e-05 (7.952 sec)
INFO:tensorflow:global_step/sec: 12.7417
INFO:tensorflow:loss = 1.2406178, step = 199900 (7.850 sec)
INFO:tensorflow:lr = 5.3645854e-05 (7.849 sec)
INFO:tensorflow:global_step/sec: 12.8213
INFO:tensorflow:loss = 1.24702, step = 200000 (7.796 sec)
INFO:tensorflow:lr = 5.3591964e-05 (7.798 sec)
INFO:tensorflow:global_step/sec: 12.8796
INFO:tensorflow:loss = 1.2420019, step = 200100 (7.768 sec)
INFO:tensorflow:lr = 5.353813e-05 (7.768 sec)
INFO:tensorflow:global_step/sec: 12.8989
INFO:tensorflow:loss = 1.242649, step = 200200 (7.758 sec)
INFO:tensorflow:lr = 5.348435e-05 (7.756 sec)
INFO:tensorflow:global_step/sec: 12.8297
INFO:tensorflow:loss = 1.2412878, step = 200300 (7.793 sec)
INFO:tensorflow:lr = 5.3430627e-05 (7.793 sec)
INFO:tensorflow:global_step/sec: 12.7035
INFO:tensorflow:loss = 1.2421048, step = 200400 (7.865 sec)
INFO:tensorflow:lr = 5.3376953e-05 (7.868 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.50905
INFO:tensorflow:loss = 1.2434052, step = 200500 (11.754 sec)
INFO:tensorflow:lr = 5.332333e-05 (11.750 sec)
INFO:tensorflow:global_step/sec: 12.6036
INFO:tensorflow:loss = 1.2436012, step = 200600 (7.937 sec)
INFO:tensorflow:lr = 5.3269774e-05 (7.938 sec)
INFO:tensorflow:global_step/sec: 12.8985
INFO:tensorflow:loss = 1.2408528, step = 200700 (7.749 sec)
INFO:tensorflow:lr = 5.321626e-05 (7.748 sec)
INFO:tensorflow:global_step/sec: 12.7336
INFO:tensorflow:loss = 1.2421123, step = 200800 (7.852 sec)
INFO:tensorflow:lr = 5.316281e-05 (7.852 sec)
INFO:tensorflow:global_step/sec: 12.9262
INFO:tensorflow:loss = 1.242066, step = 200900 (7.738 sec)
INFO:tensorflow:lr = 5.31094e-05 (7.742 sec)
INFO:tensorflow:global_step/sec: 12.5781
INFO:tensorflow:loss = 1.2408648, step = 201000 (7.954 sec)
INFO:tensorflow:lr = 5.3056046e-05 (7.949 sec)
INFO:tensorflow:global_step/sec: 12.7527
INFO:tensorflow:loss = 1.2473798, step = 201100 (7.837 sec)
INFO:tensorflow:lr = 5.3002754e-05 (7.843 sec)
INFO:tensorflow:global_step/sec: 12.9835
INFO:tensorflow:loss = 1.2424769, step = 201200 (7.708 sec)
INFO:tensorflow:lr = 5.2949512e-05 (7.702 sec)
INFO:tensorflow:global_step/sec: 12.631
INFO:tensorflow:loss = 1.2482032, step = 201300 (7.916 sec)
INFO:tensorflow:lr = 5.2896325e-05 (7.915 sec)
INFO:tensorflow:global_step/sec: 12.6894
INFO:tensorflow:loss = 1.2401179, step = 201400 (7.875 sec)
INFO:tensorflow:lr = 5.284318e-05 (7.879 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.22447
INFO:tensorflow:loss = 1.2414714, step = 201500 (10.842 sec)
INFO:tensorflow:lr = 5.27901e-05 (10.838 sec)
INFO:tensorflow:global_step/sec: 12.8635
INFO:tensorflow:loss = 1.2402923, step = 201600 (7.779 sec)
INFO:tensorflow:lr = 5.2737072e-05 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.6169
INFO:tensorflow:loss = 1.2466598, step = 201700 (7.922 sec)
INFO:tensorflow:lr = 5.2684096e-05 (7.924 sec)
INFO:tensorflow:global_step/sec: 12.8282
INFO:tensorflow:loss = 1.2450514, step = 201800 (7.798 sec)
INFO:tensorflow:lr = 5.2631174e-05 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.8332
INFO:tensorflow:loss = 1.2427869, step = 201900 (7.791 sec)
INFO:tensorflow:lr = 5.2578303e-05 (7.792 sec)
INFO:tensorflow:global_step/sec: 12.8144
INFO:tensorflow:loss = 1.2411528, step = 202000 (7.804 sec)
INFO:tensorflow:lr = 5.2525484e-05 (7.805 sec)
INFO:tensorflow:global_step/sec: 12.8084
INFO:tensorflow:loss = 1.2449394, step = 202100 (7.808 sec)
INFO:tensorflow:lr = 5.2472726e-05 (7.806 sec)
INFO:tensorflow:global_step/sec: 12.7354
INFO:tensorflow:loss = 1.243877, step = 202200 (7.847 sec)
INFO:tensorflow:lr = 5.2420008e-05 (7.851 sec)
INFO:tensorflow:global_step/sec: 12.7724
INFO:tensorflow:loss = 1.243457, step = 202300 (7.833 sec)
INFO:tensorflow:lr = 5.236736e-05 (7.829 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0221
INFO:tensorflow:loss = 1.2485676, step = 202400 (8.321 sec)
INFO:tensorflow:lr = 5.231475e-05 (8.321 sec)
INFO:tensorflow:global_step/sec: 10.6252
INFO:tensorflow:loss = 1.2417319, step = 202500 (9.409 sec)
INFO:tensorflow:lr = 5.2262196e-05 (9.409 sec)
INFO:tensorflow:global_step/sec: 12.5426
INFO:tensorflow:loss = 1.2401463, step = 202600 (7.973 sec)
INFO:tensorflow:lr = 5.2209703e-05 (7.976 sec)
INFO:tensorflow:global_step/sec: 12.6906
INFO:tensorflow:loss = 1.2469463, step = 202700 (7.881 sec)
INFO:tensorflow:lr = 5.215725e-05 (7.878 sec)
INFO:tensorflow:global_step/sec: 12.7801
INFO:tensorflow:loss = 1.2424095, step = 202800 (7.823 sec)
INFO:tensorflow:lr = 5.2104864e-05 (7.824 sec)
INFO:tensorflow:global_step/sec: 12.8786
INFO:tensorflow:loss = 1.2429019, step = 202900 (7.765 sec)
INFO:tensorflow:lr = 5.205252e-05 (7.764 sec)
INFO:tensorflow:global_step/sec: 12.5697
INFO:tensorflow:loss = 1.2409177, step = 203000 (7.951 sec)
INFO:tensorflow:lr = 5.200023e-05 (7.953 sec)
INFO:tensorflow:global_step/sec: 12.8791
INFO:tensorflow:loss = 1.2402259, step = 203100 (7.771 sec)
INFO:tensorflow:lr = 5.1948005e-05 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.87
INFO:tensorflow:loss = 1.2429622, step = 203200 (7.764 sec)
INFO:tensorflow:lr = 5.1895815e-05 (7.769 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.8056
INFO:tensorflow:loss = 1.2428799, step = 203300 (7.810 sec)
INFO:tensorflow:lr = 5.1843683e-05 (7.807 sec)
INFO:tensorflow:global_step/sec: 11.0626
INFO:tensorflow:loss = 1.2453337, step = 203400 (9.044 sec)
INFO:tensorflow:lr = 5.179161e-05 (9.044 sec)
INFO:tensorflow:global_step/sec: 12.7831
INFO:tensorflow:loss = 1.2423886, step = 203500 (7.823 sec)
INFO:tensorflow:lr = 5.173957e-05 (7.821 sec)
INFO:tensorflow:global_step/sec: 12.9267
INFO:tensorflow:loss = 1.2417992, step = 203600 (7.734 sec)
INFO:tensorflow:lr = 5.1687603e-05 (7.734 sec)
INFO:tensorflow:global_step/sec: 12.8584
INFO:tensorflow:loss = 1.2422185, step = 203700 (7.774 sec)
INFO:tensorflow:lr = 5.163568e-05 (7.774 sec)
INFO:tensorflow:global_step/sec: 12.9589
INFO:tensorflow:loss = 1.2430236, step = 203800 (7.721 sec)
INFO:tensorflow:lr = 5.1583815e-05 (7.722 sec)
INFO:tensorflow:global_step/sec: 13.0881
INFO:tensorflow:loss = 1.2422028, step = 203900 (7.641 sec)
INFO:tensorflow:lr = 5.1532e-05 (7.640 sec)
INFO:tensorflow:global_step/sec: 12.7424
INFO:tensorflow:loss = 1.244065, step = 204000 (7.842 sec)
INFO:tensorflow:lr = 5.1480227e-05 (7.842 sec)
INFO:tensorflow:global_step/sec: 12.5838
INFO:tensorflow:loss = 1.2406197, step = 204100 (7.949 sec)
INFO:tensorflow:lr = 5.142852e-05 (7.949 sec)
INFO:tensorflow:global_step/sec: 13.0673
INFO:tensorflow:loss = 1.2419794, step = 204200 (7.653 sec)
INFO:tensorflow:lr = 5.1376854e-05 (7.653 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.8192
INFO:tensorflow:loss = 1.2445016, step = 204300 (8.465 sec)
INFO:tensorflow:lr = 5.132525e-05 (8.469 sec)
INFO:tensorflow:global_step/sec: 11.1892
INFO:tensorflow:loss = 1.2439507, step = 204400 (8.934 sec)
INFO:tensorflow:lr = 5.127369e-05 (8.931 sec)
INFO:tensorflow:global_step/sec: 12.8959
INFO:tensorflow:loss = 1.2410576, step = 204500 (7.754 sec)
INFO:tensorflow:lr = 5.122218e-05 (7.753 sec)
INFO:tensorflow:global_step/sec: 12.7912
INFO:tensorflow:loss = 1.2469463, step = 204600 (7.819 sec)
INFO:tensorflow:lr = 5.117073e-05 (7.819 sec)
INFO:tensorflow:global_step/sec: 12.7463
INFO:tensorflow:loss = 1.2434845, step = 204700 (7.843 sec)
INFO:tensorflow:lr = 5.1119325e-05 (7.842 sec)
INFO:tensorflow:global_step/sec: 12.874
INFO:tensorflow:loss = 1.2412622, step = 204800 (7.767 sec)
INFO:tensorflow:lr = 5.1067982e-05 (7.772 sec)
INFO:tensorflow:global_step/sec: 12.964
INFO:tensorflow:loss = 1.2421836, step = 204900 (7.714 sec)
INFO:tensorflow:lr = 5.1016676e-05 (7.708 sec)
INFO:tensorflow:global_step/sec: 12.9155
INFO:tensorflow:loss = 1.2425599, step = 205000 (7.746 sec)
INFO:tensorflow:lr = 5.0965427e-05 (7.747 sec)
INFO:tensorflow:global_step/sec: 12.5868
INFO:tensorflow:loss = 1.2451409, step = 205100 (7.941 sec)
INFO:tensorflow:lr = 5.0914234e-05 (7.941 sec)
INFO:tensorflow:global_step/sec: 12.7319
INFO:tensorflow:loss = 1.2429577, step = 205200 (7.858 sec)
INFO:tensorflow:lr = 5.0863087e-05 (7.859 sec)
INFO:tensorflow:global_step/sec: 12.7461
INFO:tensorflow:loss = 1.2407069, step = 205300 (7.842 sec)
INFO:tensorflow:lr = 5.0812e-05 (7.844 sec)
INFO:tensorflow:Saving checkpoints for 205380 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2441876.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-205380
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-205380
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.729
INFO:tensorflow:Best (Exact Match) Accuracy: 0.733
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-205380
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 205380 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2418262, step = 205380
INFO:tensorflow:lr = 5.077116e-05
INFO:tensorflow:global_step/sec: 10.4309
INFO:tensorflow:loss = 1.243347, step = 205480 (9.594 sec)
INFO:tensorflow:lr = 5.0720155e-05 (9.593 sec)
INFO:tensorflow:global_step/sec: 12.4196
INFO:tensorflow:loss = 1.2491117, step = 205580 (8.050 sec)
INFO:tensorflow:lr = 5.0669198e-05 (8.046 sec)
INFO:tensorflow:global_step/sec: 12.5694
INFO:tensorflow:loss = 1.2417834, step = 205680 (7.960 sec)
INFO:tensorflow:lr = 5.0618306e-05 (7.960 sec)
INFO:tensorflow:global_step/sec: 12.727
INFO:tensorflow:loss = 1.2427137, step = 205780 (7.854 sec)
INFO:tensorflow:lr = 5.0567458e-05 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.9172
INFO:tensorflow:loss = 1.2439225, step = 205880 (7.743 sec)
INFO:tensorflow:lr = 5.0516664e-05 (7.742 sec)
INFO:tensorflow:global_step/sec: 12.8616
INFO:tensorflow:loss = 1.2396872, step = 205980 (7.769 sec)
INFO:tensorflow:lr = 5.0465915e-05 (7.769 sec)
INFO:tensorflow:global_step/sec: 12.8284
INFO:tensorflow:loss = 1.2469605, step = 206080 (7.800 sec)
INFO:tensorflow:lr = 5.0415223e-05 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.9143
INFO:tensorflow:loss = 1.2422843, step = 206180 (7.743 sec)
INFO:tensorflow:lr = 5.0364582e-05 (7.745 sec)
INFO:tensorflow:global_step/sec: 12.8145
INFO:tensorflow:loss = 1.2406392, step = 206280 (7.799 sec)
INFO:tensorflow:lr = 5.031399e-05 (7.798 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.21253
INFO:tensorflow:loss = 1.2431515, step = 206380 (12.183 sec)
INFO:tensorflow:lr = 5.026345e-05 (12.182 sec)
INFO:tensorflow:global_step/sec: 12.6628
INFO:tensorflow:loss = 1.2423357, step = 206480 (7.895 sec)
INFO:tensorflow:lr = 5.0212955e-05 (7.895 sec)
INFO:tensorflow:global_step/sec: 12.9072
INFO:tensorflow:loss = 1.2458041, step = 206580 (7.750 sec)
INFO:tensorflow:lr = 5.016251e-05 (7.752 sec)
INFO:tensorflow:global_step/sec: 12.6023
INFO:tensorflow:loss = 1.2403214, step = 206680 (7.929 sec)
INFO:tensorflow:lr = 5.011213e-05 (7.927 sec)
INFO:tensorflow:global_step/sec: 12.7342
INFO:tensorflow:loss = 1.24217, step = 206780 (7.857 sec)
INFO:tensorflow:lr = 5.006179e-05 (7.858 sec)
INFO:tensorflow:global_step/sec: 12.8234
INFO:tensorflow:loss = 1.2415317, step = 206880 (7.800 sec)
INFO:tensorflow:lr = 5.0011495e-05 (7.799 sec)
INFO:tensorflow:global_step/sec: 12.7469
INFO:tensorflow:loss = 1.2417748, step = 206980 (7.845 sec)
INFO:tensorflow:lr = 4.9961258e-05 (7.845 sec)
INFO:tensorflow:global_step/sec: 12.7289
INFO:tensorflow:loss = 1.2420164, step = 207080 (7.855 sec)
INFO:tensorflow:lr = 4.991107e-05 (7.855 sec)
INFO:tensorflow:global_step/sec: 12.216
INFO:tensorflow:loss = 1.242654, step = 207180 (8.184 sec)
INFO:tensorflow:lr = 4.9860933e-05 (8.184 sec)
INFO:tensorflow:global_step/sec: 13.0354
INFO:tensorflow:loss = 1.241695, step = 207280 (7.668 sec)
INFO:tensorflow:lr = 4.9810842e-05 (7.668 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.23015
INFO:tensorflow:loss = 1.2435169, step = 207380 (12.156 sec)
INFO:tensorflow:lr = 4.9760813e-05 (12.156 sec)
INFO:tensorflow:global_step/sec: 12.9586
INFO:tensorflow:loss = 1.2414528, step = 207480 (7.718 sec)
INFO:tensorflow:lr = 4.9710827e-05 (7.718 sec)
INFO:tensorflow:global_step/sec: 12.6373
INFO:tensorflow:loss = 1.2421968, step = 207580 (7.911 sec)
INFO:tensorflow:lr = 4.9660885e-05 (7.912 sec)
INFO:tensorflow:global_step/sec: 12.6402
INFO:tensorflow:loss = 1.2409348, step = 207680 (7.906 sec)
INFO:tensorflow:lr = 4.9611004e-05 (7.906 sec)
INFO:tensorflow:global_step/sec: 12.7392
INFO:tensorflow:loss = 1.2412691, step = 207780 (7.854 sec)
INFO:tensorflow:lr = 4.9561164e-05 (7.854 sec)
INFO:tensorflow:global_step/sec: 12.943
INFO:tensorflow:loss = 1.244255, step = 207880 (7.722 sec)
INFO:tensorflow:lr = 4.9511385e-05 (7.725 sec)
INFO:tensorflow:global_step/sec: 12.7753
INFO:tensorflow:loss = 1.2425479, step = 207980 (7.832 sec)
INFO:tensorflow:lr = 4.9461647e-05 (7.829 sec)
INFO:tensorflow:global_step/sec: 12.9729
INFO:tensorflow:loss = 1.2428502, step = 208080 (7.704 sec)
INFO:tensorflow:lr = 4.9411956e-05 (7.704 sec)
INFO:tensorflow:global_step/sec: 12.7115
INFO:tensorflow:loss = 1.2472405, step = 208180 (7.868 sec)
INFO:tensorflow:lr = 4.9362327e-05 (7.868 sec)
INFO:tensorflow:global_step/sec: 12.7367
INFO:tensorflow:loss = 1.2415825, step = 208280 (7.851 sec)
INFO:tensorflow:lr = 4.931274e-05 (7.851 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.47069
INFO:tensorflow:loss = 1.2403713, step = 208380 (11.811 sec)
INFO:tensorflow:lr = 4.9263206e-05 (11.811 sec)
INFO:tensorflow:global_step/sec: 12.6213
INFO:tensorflow:loss = 1.2408, step = 208480 (7.922 sec)
INFO:tensorflow:lr = 4.921372e-05 (7.921 sec)
INFO:tensorflow:global_step/sec: 12.8692
INFO:tensorflow:loss = 1.2405838, step = 208580 (7.770 sec)
INFO:tensorflow:lr = 4.9164275e-05 (7.770 sec)
INFO:tensorflow:global_step/sec: 12.9359
INFO:tensorflow:loss = 1.2444116, step = 208680 (7.726 sec)
INFO:tensorflow:lr = 4.9114893e-05 (7.726 sec)
INFO:tensorflow:global_step/sec: 12.8071
INFO:tensorflow:loss = 1.2422515, step = 208780 (7.808 sec)
INFO:tensorflow:lr = 4.906555e-05 (7.812 sec)
INFO:tensorflow:global_step/sec: 12.4937
INFO:tensorflow:loss = 1.240635, step = 208880 (8.004 sec)
INFO:tensorflow:lr = 4.9016267e-05 (8.001 sec)
INFO:tensorflow:global_step/sec: 12.9984
INFO:tensorflow:loss = 1.2395564, step = 208980 (7.698 sec)
INFO:tensorflow:lr = 4.896703e-05 (7.696 sec)
INFO:tensorflow:global_step/sec: 12.6199
INFO:tensorflow:loss = 1.2414572, step = 209080 (7.924 sec)
INFO:tensorflow:lr = 4.8917842e-05 (7.925 sec)
INFO:tensorflow:global_step/sec: 12.6925
INFO:tensorflow:loss = 1.2428277, step = 209180 (7.880 sec)
INFO:tensorflow:lr = 4.8868704e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.7822
INFO:tensorflow:loss = 1.2430718, step = 209280 (7.825 sec)
INFO:tensorflow:lr = 4.8819613e-05 (7.825 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.37639
INFO:tensorflow:loss = 1.2407434, step = 209380 (11.932 sec)
INFO:tensorflow:lr = 4.8770577e-05 (11.932 sec)
INFO:tensorflow:global_step/sec: 12.6544
INFO:tensorflow:loss = 1.2410898, step = 209480 (7.902 sec)
INFO:tensorflow:lr = 4.8721577e-05 (7.902 sec)
INFO:tensorflow:global_step/sec: 12.7381
INFO:tensorflow:loss = 1.2487592, step = 209580 (7.856 sec)
INFO:tensorflow:lr = 4.8672635e-05 (7.855 sec)
INFO:tensorflow:global_step/sec: 12.5663
INFO:tensorflow:loss = 1.2414043, step = 209680 (7.959 sec)
INFO:tensorflow:lr = 4.8623744e-05 (7.960 sec)
INFO:tensorflow:global_step/sec: 12.5939
INFO:tensorflow:loss = 1.2406616, step = 209780 (7.936 sec)
INFO:tensorflow:lr = 4.85749e-05 (7.935 sec)
INFO:tensorflow:global_step/sec: 12.5459
INFO:tensorflow:loss = 1.2446214, step = 209880 (7.970 sec)
INFO:tensorflow:lr = 4.852611e-05 (7.971 sec)
INFO:tensorflow:global_step/sec: 12.5481
INFO:tensorflow:loss = 1.2434163, step = 209980 (7.968 sec)
INFO:tensorflow:lr = 4.8477363e-05 (7.968 sec)
INFO:tensorflow:global_step/sec: 12.76
INFO:tensorflow:loss = 1.2430301, step = 210080 (7.837 sec)
INFO:tensorflow:lr = 4.842866e-05 (7.837 sec)
INFO:tensorflow:global_step/sec: 12.5294
INFO:tensorflow:loss = 1.2408226, step = 210180 (7.987 sec)
INFO:tensorflow:lr = 4.8380018e-05 (7.987 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.5734
INFO:tensorflow:loss = 1.2454717, step = 210280 (11.673 sec)
INFO:tensorflow:lr = 4.8331414e-05 (11.673 sec)
INFO:tensorflow:global_step/sec: 12.6821
INFO:tensorflow:loss = 1.2445272, step = 210380 (7.877 sec)
INFO:tensorflow:lr = 4.828287e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.5257
INFO:tensorflow:loss = 1.2420585, step = 210480 (7.977 sec)
INFO:tensorflow:lr = 4.8234368e-05 (7.976 sec)
INFO:tensorflow:global_step/sec: 12.6337
INFO:tensorflow:loss = 1.2436208, step = 210580 (7.915 sec)
INFO:tensorflow:lr = 4.818591e-05 (7.915 sec)
INFO:tensorflow:global_step/sec: 12.8119
INFO:tensorflow:loss = 1.2423323, step = 210680 (7.811 sec)
INFO:tensorflow:lr = 4.813751e-05 (7.810 sec)
INFO:tensorflow:global_step/sec: 12.4684
INFO:tensorflow:loss = 1.2410481, step = 210780 (8.018 sec)
INFO:tensorflow:lr = 4.808915e-05 (8.018 sec)
INFO:tensorflow:global_step/sec: 12.8201
INFO:tensorflow:loss = 1.2407342, step = 210880 (7.797 sec)
INFO:tensorflow:lr = 4.804085e-05 (7.797 sec)
INFO:tensorflow:global_step/sec: 12.5921
INFO:tensorflow:loss = 1.2401924, step = 210980 (7.944 sec)
INFO:tensorflow:lr = 4.7992587e-05 (7.945 sec)
INFO:tensorflow:global_step/sec: 12.7187
INFO:tensorflow:loss = 1.2426429, step = 211080 (7.862 sec)
INFO:tensorflow:lr = 4.7944373e-05 (7.862 sec)
INFO:tensorflow:global_step/sec: 12.6319
INFO:tensorflow:loss = 1.2419528, step = 211180 (7.915 sec)
INFO:tensorflow:lr = 4.7896217e-05 (7.915 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.23778
INFO:tensorflow:loss = 1.2429737, step = 211280 (10.830 sec)
INFO:tensorflow:lr = 4.7848105e-05 (10.831 sec)
INFO:tensorflow:global_step/sec: 12.8513
INFO:tensorflow:loss = 1.2444791, step = 211380 (7.780 sec)
INFO:tensorflow:lr = 4.7800044e-05 (7.780 sec)
INFO:tensorflow:global_step/sec: 12.8092
INFO:tensorflow:loss = 1.2424067, step = 211480 (7.810 sec)
INFO:tensorflow:lr = 4.7752023e-05 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.8037
INFO:tensorflow:loss = 1.2412364, step = 211580 (7.805 sec)
INFO:tensorflow:lr = 4.770405e-05 (7.809 sec)
INFO:tensorflow:global_step/sec: 12.5397
INFO:tensorflow:loss = 1.2432137, step = 211680 (7.979 sec)
INFO:tensorflow:lr = 4.7656133e-05 (7.975 sec)
INFO:tensorflow:global_step/sec: 13.0206
INFO:tensorflow:loss = 1.2420354, step = 211780 (7.676 sec)
INFO:tensorflow:lr = 4.760826e-05 (7.677 sec)
INFO:tensorflow:global_step/sec: 12.9015
INFO:tensorflow:loss = 1.2408917, step = 211880 (7.751 sec)
INFO:tensorflow:lr = 4.7560436e-05 (7.751 sec)
INFO:tensorflow:global_step/sec: 12.671
INFO:tensorflow:loss = 1.24008, step = 211980 (7.889 sec)
INFO:tensorflow:lr = 4.751266e-05 (7.889 sec)
INFO:tensorflow:global_step/sec: 12.6053
INFO:tensorflow:loss = 1.2407787, step = 212080 (7.933 sec)
INFO:tensorflow:lr = 4.746493e-05 (7.933 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1759
INFO:tensorflow:loss = 1.240133, step = 212180 (8.213 sec)
INFO:tensorflow:lr = 4.7417256e-05 (8.220 sec)
INFO:tensorflow:global_step/sec: 10.5411
INFO:tensorflow:loss = 1.242667, step = 212280 (9.488 sec)
INFO:tensorflow:lr = 4.7369624e-05 (9.481 sec)
INFO:tensorflow:global_step/sec: 12.9632
INFO:tensorflow:loss = 1.2408278, step = 212380 (7.713 sec)
INFO:tensorflow:lr = 4.732204e-05 (7.718 sec)
INFO:tensorflow:global_step/sec: 12.693
INFO:tensorflow:loss = 1.2441196, step = 212480 (7.878 sec)
INFO:tensorflow:lr = 4.7274498e-05 (7.873 sec)
INFO:tensorflow:global_step/sec: 12.7106
INFO:tensorflow:loss = 1.242603, step = 212580 (7.874 sec)
INFO:tensorflow:lr = 4.722701e-05 (7.873 sec)
INFO:tensorflow:global_step/sec: 12.5065
INFO:tensorflow:loss = 1.2410915, step = 212680 (7.990 sec)
INFO:tensorflow:lr = 4.7179576e-05 (7.991 sec)
INFO:tensorflow:global_step/sec: 12.6562
INFO:tensorflow:loss = 1.2419416, step = 212780 (7.907 sec)
INFO:tensorflow:lr = 4.7132176e-05 (7.906 sec)
INFO:tensorflow:global_step/sec: 12.3483
INFO:tensorflow:loss = 1.2452627, step = 212880 (8.097 sec)
INFO:tensorflow:lr = 4.708484e-05 (8.097 sec)
INFO:tensorflow:global_step/sec: 12.5276
INFO:tensorflow:loss = 1.2399526, step = 212980 (7.982 sec)
INFO:tensorflow:lr = 4.7037538e-05 (7.982 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.5239
INFO:tensorflow:loss = 1.2448621, step = 213080 (7.982 sec)
INFO:tensorflow:lr = 4.6990284e-05 (7.982 sec)
INFO:tensorflow:global_step/sec: 10.8555
INFO:tensorflow:loss = 1.2401872, step = 213180 (9.216 sec)
INFO:tensorflow:lr = 4.6943085e-05 (9.216 sec)
INFO:tensorflow:global_step/sec: 12.6474
INFO:tensorflow:loss = 1.2426721, step = 213280 (7.907 sec)
INFO:tensorflow:lr = 4.6895926e-05 (7.907 sec)
INFO:tensorflow:global_step/sec: 13.0417
INFO:tensorflow:loss = 1.2432199, step = 213380 (7.663 sec)
INFO:tensorflow:lr = 4.684882e-05 (7.665 sec)
INFO:tensorflow:global_step/sec: 12.7907
INFO:tensorflow:loss = 1.2436359, step = 213480 (7.821 sec)
INFO:tensorflow:lr = 4.6801757e-05 (7.820 sec)
INFO:tensorflow:global_step/sec: 12.6265
INFO:tensorflow:loss = 1.239592, step = 213580 (7.915 sec)
INFO:tensorflow:lr = 4.6754736e-05 (7.915 sec)
INFO:tensorflow:global_step/sec: 11.9851
INFO:tensorflow:loss = 1.2443177, step = 213680 (8.343 sec)
INFO:tensorflow:lr = 4.6707773e-05 (8.343 sec)
INFO:tensorflow:global_step/sec: 12.5796
INFO:tensorflow:loss = 1.2432239, step = 213780 (7.951 sec)
INFO:tensorflow:lr = 4.666086e-05 (7.951 sec)
INFO:tensorflow:global_step/sec: 12.4314
INFO:tensorflow:loss = 1.2420282, step = 213880 (8.047 sec)
INFO:tensorflow:lr = 4.661399e-05 (8.047 sec)
INFO:tensorflow:global_step/sec: 12.4872
INFO:tensorflow:loss = 1.2413042, step = 213980 (8.008 sec)
INFO:tensorflow:lr = 4.6567162e-05 (8.007 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.6035
INFO:tensorflow:loss = 1.2453946, step = 214080 (8.632 sec)
INFO:tensorflow:lr = 4.6520378e-05 (8.633 sec)
INFO:tensorflow:global_step/sec: 11.1427
INFO:tensorflow:loss = 1.241159, step = 214180 (8.963 sec)
INFO:tensorflow:lr = 4.647365e-05 (8.962 sec)
INFO:tensorflow:global_step/sec: 12.6876
INFO:tensorflow:loss = 1.240063, step = 214280 (7.880 sec)
INFO:tensorflow:lr = 4.642696e-05 (7.881 sec)
INFO:tensorflow:global_step/sec: 12.1038
INFO:tensorflow:loss = 1.2410262, step = 214380 (8.263 sec)
INFO:tensorflow:lr = 4.638033e-05 (8.263 sec)
INFO:tensorflow:global_step/sec: 12.6095
INFO:tensorflow:loss = 1.2421927, step = 214480 (7.929 sec)
INFO:tensorflow:lr = 4.6333746e-05 (7.932 sec)
INFO:tensorflow:global_step/sec: 12.2797
INFO:tensorflow:loss = 1.2429354, step = 214580 (8.144 sec)
INFO:tensorflow:lr = 4.6287194e-05 (8.142 sec)
INFO:tensorflow:global_step/sec: 12.4471
INFO:tensorflow:loss = 1.2444017, step = 214680 (8.029 sec)
INFO:tensorflow:lr = 4.6240704e-05 (8.028 sec)
INFO:tensorflow:global_step/sec: 12.5796
INFO:tensorflow:loss = 1.2418594, step = 214780 (7.949 sec)
INFO:tensorflow:lr = 4.6194247e-05 (7.953 sec)
INFO:tensorflow:global_step/sec: 12.1706
INFO:tensorflow:loss = 1.2411361, step = 214880 (8.217 sec)
INFO:tensorflow:lr = 4.614785e-05 (8.213 sec)
INFO:tensorflow:global_step/sec: 12.3648
INFO:tensorflow:loss = 1.2421589, step = 214980 (8.087 sec)
INFO:tensorflow:lr = 4.610149e-05 (8.087 sec)
INFO:tensorflow:global_step/sec: 12.6355
INFO:tensorflow:loss = 1.240195, step = 215080 (7.914 sec)
INFO:tensorflow:lr = 4.605517e-05 (7.915 sec)
INFO:tensorflow:Saving checkpoints for 215160 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2434119.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-215160
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-215160
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.731
INFO:tensorflow:Best (Exact Match) Accuracy: 0.733
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-215160
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 215160 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.2435311, step = 215160
INFO:tensorflow:lr = 4.601817e-05
INFO:tensorflow:global_step/sec: 9.90966
INFO:tensorflow:loss = 1.242061, step = 215260 (10.099 sec)
INFO:tensorflow:lr = 4.5971938e-05 (10.099 sec)
INFO:tensorflow:global_step/sec: 12.3753
INFO:tensorflow:loss = 1.2418321, step = 215360 (8.078 sec)
INFO:tensorflow:lr = 4.5925754e-05 (8.078 sec)
INFO:tensorflow:global_step/sec: 12.8241
INFO:tensorflow:loss = 1.2416762, step = 215460 (7.797 sec)
INFO:tensorflow:lr = 4.5879624e-05 (7.796 sec)
INFO:tensorflow:global_step/sec: 12.4313
INFO:tensorflow:loss = 1.2419561, step = 215560 (8.047 sec)
INFO:tensorflow:lr = 4.5833534e-05 (8.047 sec)
INFO:tensorflow:global_step/sec: 12.3863
INFO:tensorflow:loss = 1.2430488, step = 215660 (8.068 sec)
INFO:tensorflow:lr = 4.5787496e-05 (8.068 sec)
INFO:tensorflow:global_step/sec: 12.0618
INFO:tensorflow:loss = 1.2430937, step = 215760 (8.296 sec)
INFO:tensorflow:lr = 4.5741504e-05 (8.295 sec)
INFO:tensorflow:global_step/sec: 12.3146
INFO:tensorflow:loss = 1.2398549, step = 215860 (8.119 sec)
INFO:tensorflow:lr = 4.569555e-05 (8.119 sec)
INFO:tensorflow:global_step/sec: 12.3494
INFO:tensorflow:loss = 1.2440462, step = 215960 (8.097 sec)
INFO:tensorflow:lr = 4.5649653e-05 (8.098 sec)
INFO:tensorflow:global_step/sec: 11.9267
INFO:tensorflow:loss = 1.242953, step = 216060 (8.387 sec)
INFO:tensorflow:lr = 4.5603792e-05 (8.386 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.87687
INFO:tensorflow:loss = 1.2415045, step = 216160 (12.694 sec)
INFO:tensorflow:lr = 4.5557987e-05 (12.695 sec)
INFO:tensorflow:global_step/sec: 12.3866
INFO:tensorflow:loss = 1.2421069, step = 216260 (8.074 sec)
INFO:tensorflow:lr = 4.551222e-05 (8.073 sec)
INFO:tensorflow:global_step/sec: 12.3103
INFO:tensorflow:loss = 1.2413859, step = 216360 (8.119 sec)
INFO:tensorflow:lr = 4.54665e-05 (8.119 sec)
INFO:tensorflow:global_step/sec: 12.2992
INFO:tensorflow:loss = 1.2421327, step = 216460 (8.135 sec)
INFO:tensorflow:lr = 4.542083e-05 (8.135 sec)
INFO:tensorflow:global_step/sec: 12.0773
INFO:tensorflow:loss = 1.2416039, step = 216560 (8.275 sec)
INFO:tensorflow:lr = 4.5375204e-05 (8.276 sec)
INFO:tensorflow:global_step/sec: 12.3892
INFO:tensorflow:loss = 1.240614, step = 216660 (8.071 sec)
INFO:tensorflow:lr = 4.5329627e-05 (8.070 sec)
INFO:tensorflow:global_step/sec: 12.207
INFO:tensorflow:loss = 1.2392285, step = 216760 (8.192 sec)
INFO:tensorflow:lr = 4.5284087e-05 (8.192 sec)
INFO:tensorflow:global_step/sec: 12.5466
INFO:tensorflow:loss = 1.2417903, step = 216860 (7.970 sec)
INFO:tensorflow:lr = 4.5238594e-05 (7.970 sec)
INFO:tensorflow:global_step/sec: 12.5423
INFO:tensorflow:loss = 1.2406605, step = 216960 (7.978 sec)
INFO:tensorflow:lr = 4.5193156e-05 (7.978 sec)
INFO:tensorflow:global_step/sec: 12.2379
INFO:tensorflow:loss = 1.2422897, step = 217060 (8.168 sec)
INFO:tensorflow:lr = 4.5147754e-05 (8.168 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.96771
INFO:tensorflow:loss = 1.2423333, step = 217160 (12.550 sec)
INFO:tensorflow:lr = 4.5102406e-05 (12.552 sec)
INFO:tensorflow:global_step/sec: 12.0423
INFO:tensorflow:loss = 1.2417514, step = 217260 (8.305 sec)
INFO:tensorflow:lr = 4.5057095e-05 (8.306 sec)
INFO:tensorflow:global_step/sec: 12.3714
INFO:tensorflow:loss = 1.2436274, step = 217360 (8.085 sec)
INFO:tensorflow:lr = 4.5011835e-05 (8.092 sec)
INFO:tensorflow:global_step/sec: 12.0762
INFO:tensorflow:loss = 1.2410803, step = 217460 (8.278 sec)
INFO:tensorflow:lr = 4.4966622e-05 (8.273 sec)
INFO:tensorflow:global_step/sec: 12.4341
INFO:tensorflow:loss = 1.2418171, step = 217560 (8.047 sec)
INFO:tensorflow:lr = 4.492145e-05 (8.045 sec)
INFO:tensorflow:global_step/sec: 12.1051
INFO:tensorflow:loss = 1.2409146, step = 217660 (8.255 sec)
INFO:tensorflow:lr = 4.487633e-05 (8.257 sec)
INFO:tensorflow:global_step/sec: 12.5272
INFO:tensorflow:loss = 1.241607, step = 217760 (7.989 sec)
INFO:tensorflow:lr = 4.483125e-05 (7.986 sec)
INFO:tensorflow:global_step/sec: 12.4025
INFO:tensorflow:loss = 1.241136, step = 217860 (8.062 sec)
INFO:tensorflow:lr = 4.478621e-05 (8.062 sec)
INFO:tensorflow:global_step/sec: 12.2296
INFO:tensorflow:loss = 1.2414784, step = 217960 (8.176 sec)
INFO:tensorflow:lr = 4.4741224e-05 (8.176 sec)
INFO:tensorflow:global_step/sec: 12.5104
INFO:tensorflow:loss = 1.2465246, step = 218060 (7.991 sec)
INFO:tensorflow:lr = 4.469628e-05 (7.992 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 7.8534
INFO:tensorflow:loss = 1.2407348, step = 218160 (12.731 sec)
INFO:tensorflow:lr = 4.4651384e-05 (12.732 sec)
INFO:tensorflow:global_step/sec: 12.3249
INFO:tensorflow:loss = 1.2407353, step = 218260 (8.115 sec)
INFO:tensorflow:lr = 4.4606528e-05 (8.114 sec)
INFO:tensorflow:global_step/sec: 12.4159
INFO:tensorflow:loss = 1.2420832, step = 218360 (8.053 sec)
INFO:tensorflow:lr = 4.456172e-05 (8.051 sec)
INFO:tensorflow:global_step/sec: 11.9979
INFO:tensorflow:loss = 1.2450037, step = 218460 (8.334 sec)
INFO:tensorflow:lr = 4.4516957e-05 (8.334 sec)
INFO:tensorflow:global_step/sec: 12.2945
INFO:tensorflow:loss = 1.2435594, step = 218560 (8.134 sec)
INFO:tensorflow:lr = 4.447224e-05 (8.135 sec)
INFO:tensorflow:global_step/sec: 12.4887
INFO:tensorflow:loss = 1.2481513, step = 218660 (8.007 sec)
INFO:tensorflow:lr = 4.442757e-05 (8.006 sec)
INFO:tensorflow:global_step/sec: 12.5222
INFO:tensorflow:loss = 1.240262, step = 218760 (7.986 sec)
INFO:tensorflow:lr = 4.438293e-05 (7.986 sec)
INFO:tensorflow:global_step/sec: 12.325
INFO:tensorflow:loss = 1.2430587, step = 218860 (8.119 sec)
INFO:tensorflow:lr = 4.433835e-05 (8.120 sec)
INFO:tensorflow:global_step/sec: 12.7887
INFO:tensorflow:loss = 1.2427464, step = 218960 (7.821 sec)
INFO:tensorflow:lr = 4.4293814e-05 (7.822 sec)
INFO:tensorflow:global_step/sec: 12.9406
INFO:tensorflow:loss = 1.2418499, step = 219060 (7.726 sec)
INFO:tensorflow:lr = 4.4249315e-05 (7.724 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.25481
INFO:tensorflow:loss = 1.2473, step = 219160 (12.109 sec)
INFO:tensorflow:lr = 4.4204873e-05 (12.109 sec)
INFO:tensorflow:global_step/sec: 12.68
INFO:tensorflow:loss = 1.2424519, step = 219260 (7.891 sec)
INFO:tensorflow:lr = 4.4160464e-05 (7.892 sec)
INFO:tensorflow:global_step/sec: 12.6321
INFO:tensorflow:loss = 1.2408296, step = 219360 (7.917 sec)
INFO:tensorflow:lr = 4.4116096e-05 (7.916 sec)
INFO:tensorflow:global_step/sec: 12.2726
INFO:tensorflow:loss = 1.2401123, step = 219460 (8.148 sec)
INFO:tensorflow:lr = 4.4071785e-05 (8.148 sec)
INFO:tensorflow:global_step/sec: 12.3966
INFO:tensorflow:loss = 1.2420714, step = 219560 (8.068 sec)
INFO:tensorflow:lr = 4.4027514e-05 (8.068 sec)
INFO:tensorflow:global_step/sec: 12.2952
INFO:tensorflow:loss = 1.2455899, step = 219660 (8.133 sec)
INFO:tensorflow:lr = 4.398329e-05 (8.133 sec)
INFO:tensorflow:global_step/sec: 12.6129
INFO:tensorflow:loss = 1.2427148, step = 219760 (7.922 sec)
INFO:tensorflow:lr = 4.3939108e-05 (7.922 sec)
INFO:tensorflow:global_step/sec: 12.6133
INFO:tensorflow:loss = 1.2426178, step = 219860 (7.928 sec)
INFO:tensorflow:lr = 4.3894965e-05 (7.928 sec)
INFO:tensorflow:global_step/sec: 12.5467
INFO:tensorflow:loss = 1.241264, step = 219960 (7.974 sec)
INFO:tensorflow:lr = 4.3850876e-05 (7.975 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.39625
INFO:tensorflow:loss = 1.2407962, step = 220060 (11.908 sec)
INFO:tensorflow:lr = 4.3806824e-05 (11.909 sec)
INFO:tensorflow:global_step/sec: 12.6177
INFO:tensorflow:loss = 1.2421613, step = 220160 (7.926 sec)
INFO:tensorflow:lr = 4.3762822e-05 (7.926 sec)
INFO:tensorflow:global_step/sec: 12.7117
INFO:tensorflow:loss = 1.2414433, step = 220260 (7.866 sec)
INFO:tensorflow:lr = 4.3718857e-05 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.8279
INFO:tensorflow:loss = 1.2425734, step = 220360 (7.794 sec)
INFO:tensorflow:lr = 4.3674943e-05 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.2026
INFO:tensorflow:loss = 1.241693, step = 220460 (8.200 sec)
INFO:tensorflow:lr = 4.3631073e-05 (8.204 sec)
INFO:tensorflow:global_step/sec: 12.4323
INFO:tensorflow:loss = 1.2427572, step = 220560 (8.045 sec)
INFO:tensorflow:lr = 4.358724e-05 (8.040 sec)
INFO:tensorflow:global_step/sec: 12.5544
INFO:tensorflow:loss = 1.2428184, step = 220660 (7.962 sec)
INFO:tensorflow:lr = 4.354346e-05 (7.962 sec)
INFO:tensorflow:global_step/sec: 12.6407
INFO:tensorflow:loss = 1.2432847, step = 220760 (7.911 sec)
INFO:tensorflow:lr = 4.3499716e-05 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.7519
INFO:tensorflow:loss = 1.2409394, step = 220860 (7.847 sec)
INFO:tensorflow:lr = 4.3456017e-05 (7.847 sec)
INFO:tensorflow:global_step/sec: 12.6855
INFO:tensorflow:loss = 1.2418656, step = 220960 (7.878 sec)
INFO:tensorflow:lr = 4.341237e-05 (7.880 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.12931
INFO:tensorflow:loss = 1.2419653, step = 221060 (10.956 sec)
INFO:tensorflow:lr = 4.3368756e-05 (10.957 sec)
INFO:tensorflow:global_step/sec: 12.8768
INFO:tensorflow:loss = 1.2422274, step = 221160 (7.761 sec)
INFO:tensorflow:lr = 4.33252e-05 (7.759 sec)
INFO:tensorflow:global_step/sec: 12.5233
INFO:tensorflow:loss = 1.239835, step = 221260 (7.985 sec)
INFO:tensorflow:lr = 4.3281674e-05 (7.990 sec)
INFO:tensorflow:global_step/sec: 12.6333
INFO:tensorflow:loss = 1.2409306, step = 221360 (7.917 sec)
INFO:tensorflow:lr = 4.323819e-05 (7.915 sec)
INFO:tensorflow:global_step/sec: 12.6529
INFO:tensorflow:loss = 1.2407382, step = 221460 (7.909 sec)
INFO:tensorflow:lr = 4.319476e-05 (7.906 sec)
INFO:tensorflow:global_step/sec: 12.6227
INFO:tensorflow:loss = 1.2425914, step = 221560 (7.920 sec)
INFO:tensorflow:lr = 4.315137e-05 (7.920 sec)
INFO:tensorflow:global_step/sec: 12.7449
INFO:tensorflow:loss = 1.2465827, step = 221660 (7.845 sec)
INFO:tensorflow:lr = 4.3108023e-05 (7.845 sec)
INFO:tensorflow:global_step/sec: 12.6797
INFO:tensorflow:loss = 1.2416872, step = 221760 (7.889 sec)
INFO:tensorflow:lr = 4.306472e-05 (7.890 sec)
INFO:tensorflow:global_step/sec: 12.721
INFO:tensorflow:loss = 1.2460314, step = 221860 (7.865 sec)
INFO:tensorflow:lr = 4.3021457e-05 (7.864 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.1301
INFO:tensorflow:loss = 1.2410057, step = 221960 (8.244 sec)
INFO:tensorflow:lr = 4.297825e-05 (8.245 sec)
INFO:tensorflow:global_step/sec: 10.8417
INFO:tensorflow:loss = 1.2405419, step = 222060 (9.219 sec)
INFO:tensorflow:lr = 4.293507e-05 (9.219 sec)
INFO:tensorflow:global_step/sec: 12.6896
INFO:tensorflow:loss = 1.2431215, step = 222160 (7.875 sec)
INFO:tensorflow:lr = 4.289194e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.6973
INFO:tensorflow:loss = 1.2419086, step = 222260 (7.876 sec)
INFO:tensorflow:lr = 4.2848853e-05 (7.877 sec)
INFO:tensorflow:global_step/sec: 12.5823
INFO:tensorflow:loss = 1.2409471, step = 222360 (7.953 sec)
INFO:tensorflow:lr = 4.2805805e-05 (7.949 sec)
INFO:tensorflow:global_step/sec: 12.8736
INFO:tensorflow:loss = 1.2430838, step = 222460 (7.762 sec)
INFO:tensorflow:lr = 4.276281e-05 (7.765 sec)
INFO:tensorflow:global_step/sec: 12.5773
INFO:tensorflow:loss = 1.242637, step = 222560 (7.951 sec)
INFO:tensorflow:lr = 4.271985e-05 (7.948 sec)
INFO:tensorflow:global_step/sec: 12.883
INFO:tensorflow:loss = 1.2404941, step = 222660 (7.768 sec)
INFO:tensorflow:lr = 4.2676944e-05 (7.767 sec)
INFO:tensorflow:global_step/sec: 12.8256
INFO:tensorflow:loss = 1.2423271, step = 222760 (7.791 sec)
INFO:tensorflow:lr = 4.263407e-05 (7.796 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.3713
INFO:tensorflow:loss = 1.2421962, step = 222860 (8.084 sec)
INFO:tensorflow:lr = 4.259124e-05 (8.080 sec)
INFO:tensorflow:global_step/sec: 11.0319
INFO:tensorflow:loss = 1.2413177, step = 222960 (9.068 sec)
INFO:tensorflow:lr = 4.254846e-05 (9.072 sec)
INFO:tensorflow:global_step/sec: 12.547
INFO:tensorflow:loss = 1.2451798, step = 223060 (7.968 sec)
INFO:tensorflow:lr = 4.250572e-05 (7.963 sec)
INFO:tensorflow:global_step/sec: 12.8653
INFO:tensorflow:loss = 1.2412239, step = 223160 (7.777 sec)
INFO:tensorflow:lr = 4.2463023e-05 (7.777 sec)
INFO:tensorflow:global_step/sec: 12.6293
INFO:tensorflow:loss = 1.2409556, step = 223260 (7.917 sec)
INFO:tensorflow:lr = 4.2420364e-05 (7.918 sec)
INFO:tensorflow:global_step/sec: 12.6865
INFO:tensorflow:loss = 1.241664, step = 223360 (7.880 sec)
INFO:tensorflow:lr = 4.237775e-05 (7.879 sec)
INFO:tensorflow:global_step/sec: 12.646
INFO:tensorflow:loss = 1.241841, step = 223460 (7.911 sec)
INFO:tensorflow:lr = 4.233519e-05 (7.911 sec)
INFO:tensorflow:global_step/sec: 12.4094
INFO:tensorflow:loss = 1.2418112, step = 223560 (8.059 sec)
INFO:tensorflow:lr = 4.229266e-05 (8.059 sec)
INFO:tensorflow:global_step/sec: 12.9008
INFO:tensorflow:loss = 1.2413185, step = 223660 (7.745 sec)
INFO:tensorflow:lr = 4.2250176e-05 (7.746 sec)
INFO:tensorflow:global_step/sec: 12.9506
INFO:tensorflow:loss = 1.2488402, step = 223760 (7.725 sec)
INFO:tensorflow:lr = 4.2207732e-05 (7.725 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.7749
INFO:tensorflow:loss = 1.2431126, step = 223860 (8.496 sec)
INFO:tensorflow:lr = 4.2165328e-05 (8.502 sec)
INFO:tensorflow:global_step/sec: 11.4856
INFO:tensorflow:loss = 1.2412598, step = 223960 (8.705 sec)
INFO:tensorflow:lr = 4.2122978e-05 (8.700 sec)
INFO:tensorflow:global_step/sec: 12.8892
INFO:tensorflow:loss = 1.2441949, step = 224060 (7.757 sec)
INFO:tensorflow:lr = 4.2080665e-05 (7.756 sec)
INFO:tensorflow:global_step/sec: 12.6143
INFO:tensorflow:loss = 1.2411182, step = 224160 (7.926 sec)
INFO:tensorflow:lr = 4.203839e-05 (7.929 sec)
INFO:tensorflow:global_step/sec: 12.9576
INFO:tensorflow:loss = 1.2425789, step = 224260 (7.719 sec)
INFO:tensorflow:lr = 4.1996165e-05 (7.716 sec)
INFO:tensorflow:global_step/sec: 12.6436
INFO:tensorflow:loss = 1.2407291, step = 224360 (7.910 sec)
INFO:tensorflow:lr = 4.1953976e-05 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.9342
INFO:tensorflow:loss = 1.241793, step = 224460 (7.727 sec)
INFO:tensorflow:lr = 4.1911837e-05 (7.726 sec)
INFO:tensorflow:global_step/sec: 12.7712
INFO:tensorflow:loss = 1.2459935, step = 224560 (7.830 sec)
INFO:tensorflow:lr = 4.186973e-05 (7.838 sec)
INFO:tensorflow:global_step/sec: 12.5492
INFO:tensorflow:loss = 1.241738, step = 224660 (7.969 sec)
INFO:tensorflow:lr = 4.1827676e-05 (7.967 sec)
INFO:tensorflow:global_step/sec: 12.9812
INFO:tensorflow:loss = 1.2425698, step = 224760 (7.706 sec)
INFO:tensorflow:lr = 4.1785654e-05 (7.700 sec)
INFO:tensorflow:global_step/sec: 12.5881
INFO:tensorflow:loss = 1.2425317, step = 224860 (7.955 sec)
INFO:tensorflow:lr = 4.1743675e-05 (7.955 sec)
INFO:tensorflow:Saving checkpoints for 224940 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.24172.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-224940
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show playing ] [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
________________________|______________________________________________
| | | | sl:location
| | | | |
| | | | in:get_location
| | | | ________________|_______________
| | | sl:category_even sl:search_radius sl:location_user
| | | t | |
| | | _________|_________________ | |
what times are the nutcracker show playing near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-224940
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv
INFO:tensorflow:Evaluation: Testing (Exact Match) Accuracy: 0.728
INFO:tensorflow:Best (Exact Match) Accuracy: 0.733
INFO:tensorflow:Calling model_fn.
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:38: DeprecationWarning: elementwise comparison failed; this will raise an error in the future.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
[<tf.Variable 'Embedding/dense/kernel:0' shape=(1024, 300) dtype=float32_ref>,
<tf.Variable 'Embedding/dense/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Embedding/glove:0' shape=(8692, 300) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/kernel:0' shape=(900, 1200) dtype=float32_ref>,
<tf.Variable 'Encoder/lstm_fused_cell_1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/affine_bias:0' shape=(8692,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_0/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/self_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/kernel:0' shape=(300, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/query/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/key/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/kernel:0' shape=(600, 512) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/value/bias:0' shape=(512,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/kernel:0' shape=(512, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/encdec_attention/multihead_attention/output/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/past_poswise_ln/gamma:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/kernel:0' shape=(300, 1200) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv1/bias:0' shape=(1200,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/kernel:0' shape=(1200, 300) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/layer_1/ffn/conv2/bias:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/beta:0' shape=(300,) dtype=float32_ref>,
<tf.Variable 'Decoder/transformer_decoder/gamma:0' shape=(300,) dtype=float32_ref>]
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-224940
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Saving checkpoints for 224940 into ../model/lstm_transformer/model.ckpt.
Reading ../data/train.tsv
INFO:tensorflow:loss = 1.245515, step = 224940
INFO:tensorflow:lr = 4.171013e-05
INFO:tensorflow:global_step/sec: 10.21
INFO:tensorflow:loss = 1.2412355, step = 225040 (9.801 sec)
INFO:tensorflow:lr = 4.166823e-05 (9.802 sec)
INFO:tensorflow:global_step/sec: 12.5524
INFO:tensorflow:loss = 1.2411782, step = 225140 (7.966 sec)
INFO:tensorflow:lr = 4.1626372e-05 (7.966 sec)
INFO:tensorflow:global_step/sec: 12.9635
INFO:tensorflow:loss = 1.2436695, step = 225240 (7.714 sec)
INFO:tensorflow:lr = 4.158456e-05 (7.714 sec)
INFO:tensorflow:global_step/sec: 12.8617
INFO:tensorflow:loss = 1.2410346, step = 225340 (7.770 sec)
INFO:tensorflow:lr = 4.1542786e-05 (7.773 sec)
INFO:tensorflow:global_step/sec: 12.7585
INFO:tensorflow:loss = 1.2414058, step = 225440 (7.842 sec)
INFO:tensorflow:lr = 4.1501055e-05 (7.840 sec)
INFO:tensorflow:global_step/sec: 12.5799
INFO:tensorflow:loss = 1.2416364, step = 225540 (7.948 sec)
INFO:tensorflow:lr = 4.145937e-05 (7.948 sec)
INFO:tensorflow:global_step/sec: 12.6295
INFO:tensorflow:loss = 1.241216, step = 225640 (7.919 sec)
INFO:tensorflow:lr = 4.141772e-05 (7.919 sec)
INFO:tensorflow:global_step/sec: 12.63
INFO:tensorflow:loss = 1.2452391, step = 225740 (7.920 sec)
INFO:tensorflow:lr = 4.137612e-05 (7.920 sec)
INFO:tensorflow:global_step/sec: 12.7772
INFO:tensorflow:loss = 1.2397666, step = 225840 (7.825 sec)
INFO:tensorflow:lr = 4.133455e-05 (7.825 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.1808
INFO:tensorflow:loss = 1.240168, step = 225940 (12.223 sec)
INFO:tensorflow:lr = 4.1293028e-05 (12.223 sec)
INFO:tensorflow:global_step/sec: 12.7142
INFO:tensorflow:loss = 1.242873, step = 226040 (7.866 sec)
INFO:tensorflow:lr = 4.125155e-05 (7.865 sec)
INFO:tensorflow:global_step/sec: 12.8279
INFO:tensorflow:loss = 1.2412157, step = 226140 (7.794 sec)
INFO:tensorflow:lr = 4.1210107e-05 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.566
INFO:tensorflow:loss = 1.2413632, step = 226240 (7.955 sec)
INFO:tensorflow:lr = 4.1168714e-05 (7.954 sec)
INFO:tensorflow:global_step/sec: 12.7163
INFO:tensorflow:loss = 1.2435849, step = 226340 (7.866 sec)
INFO:tensorflow:lr = 4.112736e-05 (7.867 sec)
INFO:tensorflow:global_step/sec: 12.8243
INFO:tensorflow:loss = 1.2454369, step = 226440 (7.796 sec)
INFO:tensorflow:lr = 4.108604e-05 (7.795 sec)
INFO:tensorflow:global_step/sec: 12.6798
INFO:tensorflow:loss = 1.245144, step = 226540 (7.887 sec)
INFO:tensorflow:lr = 4.1044772e-05 (7.886 sec)
INFO:tensorflow:global_step/sec: 12.7597
INFO:tensorflow:loss = 1.2447907, step = 226640 (7.837 sec)
INFO:tensorflow:lr = 4.100354e-05 (7.837 sec)
INFO:tensorflow:global_step/sec: 12.6933
INFO:tensorflow:loss = 1.2408261, step = 226740 (7.882 sec)
INFO:tensorflow:lr = 4.0962354e-05 (7.883 sec)
INFO:tensorflow:global_step/sec: 12.4884
INFO:tensorflow:loss = 1.240148, step = 226840 (8.009 sec)
INFO:tensorflow:lr = 4.09212e-05 (8.008 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.39542
INFO:tensorflow:loss = 1.2397572, step = 226940 (11.906 sec)
INFO:tensorflow:lr = 4.0880095e-05 (11.906 sec)
INFO:tensorflow:global_step/sec: 12.7057
INFO:tensorflow:loss = 1.2418107, step = 227040 (7.877 sec)
INFO:tensorflow:lr = 4.0839033e-05 (7.877 sec)
INFO:tensorflow:global_step/sec: 12.6331
INFO:tensorflow:loss = 1.2409166, step = 227140 (7.910 sec)
INFO:tensorflow:lr = 4.0798008e-05 (7.910 sec)
INFO:tensorflow:global_step/sec: 12.865
INFO:tensorflow:loss = 1.2418015, step = 227240 (7.772 sec)
INFO:tensorflow:lr = 4.075703e-05 (7.772 sec)
INFO:tensorflow:global_step/sec: 12.8334
INFO:tensorflow:loss = 1.2457359, step = 227340 (7.797 sec)
INFO:tensorflow:lr = 4.0716084e-05 (7.797 sec)
INFO:tensorflow:global_step/sec: 12.8722
INFO:tensorflow:loss = 1.2415484, step = 227440 (7.768 sec)
INFO:tensorflow:lr = 4.067518e-05 (7.768 sec)
INFO:tensorflow:global_step/sec: 12.6562
INFO:tensorflow:loss = 1.2438579, step = 227540 (7.902 sec)
INFO:tensorflow:lr = 4.0634328e-05 (7.902 sec)
INFO:tensorflow:global_step/sec: 12.728
INFO:tensorflow:loss = 1.2444311, step = 227640 (7.854 sec)
INFO:tensorflow:lr = 4.0593506e-05 (7.858 sec)
INFO:tensorflow:global_step/sec: 12.7264
INFO:tensorflow:loss = 1.2432747, step = 227740 (7.860 sec)
INFO:tensorflow:lr = 4.055273e-05 (7.855 sec)
INFO:tensorflow:global_step/sec: 12.9611
INFO:tensorflow:loss = 1.2415364, step = 227840 (7.717 sec)
INFO:tensorflow:lr = 4.0511994e-05 (7.718 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.30722
INFO:tensorflow:loss = 1.2399315, step = 227940 (12.039 sec)
INFO:tensorflow:lr = 4.0471296e-05 (12.038 sec)
INFO:tensorflow:global_step/sec: 12.6838
INFO:tensorflow:loss = 1.2412345, step = 228040 (7.878 sec)
INFO:tensorflow:lr = 4.0430645e-05 (7.878 sec)
INFO:tensorflow:global_step/sec: 12.9424
INFO:tensorflow:loss = 1.2414714, step = 228140 (7.732 sec)
INFO:tensorflow:lr = 4.039003e-05 (7.732 sec)
INFO:tensorflow:global_step/sec: 13.0548
INFO:tensorflow:loss = 1.2409471, step = 228240 (7.660 sec)
INFO:tensorflow:lr = 4.034946e-05 (7.659 sec)
INFO:tensorflow:global_step/sec: 12.6404
INFO:tensorflow:loss = 1.2413936, step = 228340 (7.906 sec)
INFO:tensorflow:lr = 4.0308925e-05 (7.907 sec)
INFO:tensorflow:global_step/sec: 12.6828
INFO:tensorflow:loss = 1.2451051, step = 228440 (7.886 sec)
INFO:tensorflow:lr = 4.026843e-05 (7.887 sec)
INFO:tensorflow:global_step/sec: 12.8877
INFO:tensorflow:loss = 1.2420641, step = 228540 (7.759 sec)
INFO:tensorflow:lr = 4.0227984e-05 (7.757 sec)
INFO:tensorflow:global_step/sec: 12.6603
INFO:tensorflow:loss = 1.2408705, step = 228640 (7.901 sec)
INFO:tensorflow:lr = 4.0187577e-05 (7.902 sec)
INFO:tensorflow:global_step/sec: 12.9905
INFO:tensorflow:loss = 1.2428869, step = 228740 (7.700 sec)
INFO:tensorflow:lr = 4.014721e-05 (7.699 sec)
INFO:tensorflow:global_step/sec: 12.8204
INFO:tensorflow:loss = 1.2405547, step = 228840 (7.800 sec)
INFO:tensorflow:lr = 4.0106876e-05 (7.803 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.38389
INFO:tensorflow:loss = 1.2480453, step = 228940 (11.929 sec)
INFO:tensorflow:lr = 4.0066585e-05 (11.926 sec)
INFO:tensorflow:global_step/sec: 12.8982
INFO:tensorflow:loss = 1.242588, step = 229040 (7.752 sec)
INFO:tensorflow:lr = 4.0026338e-05 (7.752 sec)
INFO:tensorflow:global_step/sec: 12.7452
INFO:tensorflow:loss = 1.2459991, step = 229140 (7.848 sec)
INFO:tensorflow:lr = 3.9986127e-05 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.8336
INFO:tensorflow:loss = 1.2415149, step = 229240 (7.788 sec)
INFO:tensorflow:lr = 3.994596e-05 (7.788 sec)
INFO:tensorflow:global_step/sec: 12.2859
INFO:tensorflow:loss = 1.242039, step = 229340 (8.141 sec)
INFO:tensorflow:lr = 3.9905834e-05 (8.142 sec)
INFO:tensorflow:global_step/sec: 12.5119
INFO:tensorflow:loss = 1.25192, step = 229440 (7.995 sec)
INFO:tensorflow:lr = 3.986575e-05 (7.993 sec)
INFO:tensorflow:global_step/sec: 12.7531
INFO:tensorflow:loss = 1.245667, step = 229540 (7.844 sec)
INFO:tensorflow:lr = 3.9825703e-05 (7.844 sec)
INFO:tensorflow:global_step/sec: 12.8561
INFO:tensorflow:loss = 1.2407379, step = 229640 (7.773 sec)
INFO:tensorflow:lr = 3.9785693e-05 (7.772 sec)
INFO:tensorflow:global_step/sec: 12.7172
INFO:tensorflow:loss = 1.2451239, step = 229740 (7.867 sec)
INFO:tensorflow:lr = 3.9745733e-05 (7.867 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 8.73178
INFO:tensorflow:loss = 1.2411838, step = 229840 (11.447 sec)
INFO:tensorflow:lr = 3.9705807e-05 (11.449 sec)
INFO:tensorflow:global_step/sec: 12.9561
INFO:tensorflow:loss = 1.2413689, step = 229940 (7.720 sec)
INFO:tensorflow:lr = 3.9665916e-05 (7.724 sec)
INFO:tensorflow:global_step/sec: 12.7091
INFO:tensorflow:loss = 1.2430978, step = 230040 (7.869 sec)
INFO:tensorflow:lr = 3.9626073e-05 (7.863 sec)
INFO:tensorflow:global_step/sec: 12.7369
INFO:tensorflow:loss = 1.242768, step = 230140 (7.851 sec)
INFO:tensorflow:lr = 3.958627e-05 (7.852 sec)
INFO:tensorflow:global_step/sec: 12.7465
INFO:tensorflow:loss = 1.2418132, step = 230240 (7.848 sec)
INFO:tensorflow:lr = 3.9546507e-05 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.9642
INFO:tensorflow:loss = 1.2418685, step = 230340 (7.710 sec)
INFO:tensorflow:lr = 3.950678e-05 (7.709 sec)
INFO:tensorflow:global_step/sec: 12.4698
INFO:tensorflow:loss = 1.2514341, step = 230440 (8.018 sec)
INFO:tensorflow:lr = 3.946709e-05 (8.019 sec)
INFO:tensorflow:global_step/sec: 12.4999
INFO:tensorflow:loss = 1.2429241, step = 230540 (8.008 sec)
INFO:tensorflow:lr = 3.942745e-05 (8.010 sec)
INFO:tensorflow:global_step/sec: 12.7492
INFO:tensorflow:loss = 1.2442654, step = 230640 (7.841 sec)
INFO:tensorflow:lr = 3.938784e-05 (7.839 sec)
INFO:tensorflow:global_step/sec: 12.6677
INFO:tensorflow:loss = 1.2414587, step = 230740 (7.889 sec)
INFO:tensorflow:lr = 3.9348277e-05 (7.890 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 9.14225
INFO:tensorflow:loss = 1.2435497, step = 230840 (10.951 sec)
INFO:tensorflow:lr = 3.9308747e-05 (10.950 sec)
INFO:tensorflow:global_step/sec: 12.8207
INFO:tensorflow:loss = 1.2424185, step = 230940 (7.791 sec)
INFO:tensorflow:lr = 3.926926e-05 (7.791 sec)
INFO:tensorflow:global_step/sec: 12.5554
INFO:tensorflow:loss = 1.2412125, step = 231040 (7.966 sec)
INFO:tensorflow:lr = 3.9229813e-05 (7.966 sec)
INFO:tensorflow:global_step/sec: 12.9233
INFO:tensorflow:loss = 1.2414656, step = 231140 (7.733 sec)
INFO:tensorflow:lr = 3.9190403e-05 (7.734 sec)
INFO:tensorflow:global_step/sec: 12.6232
INFO:tensorflow:loss = 1.2446767, step = 231240 (7.922 sec)
INFO:tensorflow:lr = 3.915104e-05 (7.926 sec)
INFO:tensorflow:global_step/sec: 12.8202
INFO:tensorflow:loss = 1.2399716, step = 231340 (7.804 sec)
INFO:tensorflow:lr = 3.911171e-05 (7.800 sec)
INFO:tensorflow:global_step/sec: 12.6593
INFO:tensorflow:loss = 1.2404665, step = 231440 (7.896 sec)
INFO:tensorflow:lr = 3.9072416e-05 (7.899 sec)
INFO:tensorflow:global_step/sec: 12.8141
INFO:tensorflow:loss = 1.2436392, step = 231540 (7.805 sec)
INFO:tensorflow:lr = 3.9033173e-05 (7.804 sec)
INFO:tensorflow:global_step/sec: 12.6215
INFO:tensorflow:loss = 1.2410214, step = 231640 (7.928 sec)
INFO:tensorflow:lr = 3.899396e-05 (7.925 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.0998
INFO:tensorflow:loss = 1.2402171, step = 231740 (8.273 sec)
INFO:tensorflow:lr = 3.8954797e-05 (8.274 sec)
INFO:tensorflow:global_step/sec: 10.8586
INFO:tensorflow:loss = 1.2410094, step = 231840 (9.195 sec)
INFO:tensorflow:lr = 3.8915663e-05 (9.195 sec)
INFO:tensorflow:global_step/sec: 12.9364
INFO:tensorflow:loss = 1.2405255, step = 231940 (7.729 sec)
INFO:tensorflow:lr = 3.887657e-05 (7.735 sec)
INFO:tensorflow:global_step/sec: 12.591
INFO:tensorflow:loss = 1.2420667, step = 232040 (7.948 sec)
INFO:tensorflow:lr = 3.883752e-05 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.6597
INFO:tensorflow:loss = 1.2402742, step = 232140 (7.898 sec)
INFO:tensorflow:lr = 3.8798502e-05 (7.898 sec)
INFO:tensorflow:global_step/sec: 12.7635
INFO:tensorflow:loss = 1.2407392, step = 232240 (7.837 sec)
INFO:tensorflow:lr = 3.875953e-05 (7.837 sec)
INFO:tensorflow:global_step/sec: 12.7974
INFO:tensorflow:loss = 1.2414687, step = 232340 (7.813 sec)
INFO:tensorflow:lr = 3.872059e-05 (7.813 sec)
INFO:tensorflow:global_step/sec: 12.7841
INFO:tensorflow:loss = 1.2432356, step = 232440 (7.823 sec)
INFO:tensorflow:lr = 3.8681694e-05 (7.824 sec)
INFO:tensorflow:global_step/sec: 12.6057
INFO:tensorflow:loss = 1.241169, step = 232540 (7.931 sec)
INFO:tensorflow:lr = 3.8642844e-05 (7.931 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 12.7646
INFO:tensorflow:loss = 1.2469207, step = 232640 (7.833 sec)
INFO:tensorflow:lr = 3.860402e-05 (7.833 sec)
INFO:tensorflow:global_step/sec: 10.9391
INFO:tensorflow:loss = 1.2405813, step = 232740 (9.139 sec)
INFO:tensorflow:lr = 3.8565246e-05 (9.139 sec)
INFO:tensorflow:global_step/sec: 12.9065
INFO:tensorflow:loss = 1.2419277, step = 232840 (7.753 sec)
INFO:tensorflow:lr = 3.8526505e-05 (7.753 sec)
INFO:tensorflow:global_step/sec: 12.7425
INFO:tensorflow:loss = 1.241205, step = 232940 (7.847 sec)
INFO:tensorflow:lr = 3.84878e-05 (7.848 sec)
INFO:tensorflow:global_step/sec: 12.6472
INFO:tensorflow:loss = 1.241587, step = 233040 (7.906 sec)
INFO:tensorflow:lr = 3.8449143e-05 (7.906 sec)
INFO:tensorflow:global_step/sec: 12.274
INFO:tensorflow:loss = 1.2420787, step = 233140 (8.144 sec)
INFO:tensorflow:lr = 3.8410515e-05 (8.142 sec)
INFO:tensorflow:global_step/sec: 12.1353
INFO:tensorflow:loss = 1.2417104, step = 233240 (8.247 sec)
INFO:tensorflow:lr = 3.8371934e-05 (8.248 sec)
INFO:tensorflow:global_step/sec: 12.1941
INFO:tensorflow:loss = 1.2428356, step = 233340 (8.194 sec)
INFO:tensorflow:lr = 3.833339e-05 (8.194 sec)
INFO:tensorflow:global_step/sec: 12.1589
INFO:tensorflow:loss = 1.2426999, step = 233440 (8.227 sec)
INFO:tensorflow:lr = 3.8294882e-05 (8.228 sec)
INFO:tensorflow:global_step/sec: 11.8032
INFO:tensorflow:loss = 1.240998, step = 233540 (8.472 sec)
INFO:tensorflow:lr = 3.8256414e-05 (8.473 sec)
Reading ../data/train.tsv
INFO:tensorflow:global_step/sec: 11.5196
INFO:tensorflow:loss = 1.2467922, step = 233640 (8.685 sec)
INFO:tensorflow:lr = 3.8217982e-05 (8.685 sec)
INFO:tensorflow:global_step/sec: 11.2044
INFO:tensorflow:loss = 1.239357, step = 233740 (8.920 sec)
INFO:tensorflow:lr = 3.8179598e-05 (8.917 sec)
INFO:tensorflow:global_step/sec: 12.3381
INFO:tensorflow:loss = 1.240096, step = 233840 (8.109 sec)
INFO:tensorflow:lr = 3.814124e-05 (8.109 sec)
INFO:tensorflow:global_step/sec: 12.2644
INFO:tensorflow:loss = 1.2404408, step = 233940 (8.156 sec)
INFO:tensorflow:lr = 3.8102924e-05 (8.155 sec)
INFO:tensorflow:global_step/sec: 12.5479
INFO:tensorflow:loss = 1.2412769, step = 234040 (7.964 sec)
INFO:tensorflow:lr = 3.8064652e-05 (7.965 sec)
INFO:tensorflow:global_step/sec: 12.4405
INFO:tensorflow:loss = 1.2444044, step = 234140 (8.037 sec)
INFO:tensorflow:lr = 3.8026414e-05 (8.040 sec)
INFO:tensorflow:global_step/sec: 12.5963
INFO:tensorflow:loss = 1.2408541, step = 234240 (7.939 sec)
INFO:tensorflow:lr = 3.798822e-05 (7.935 sec)
INFO:tensorflow:global_step/sec: 12.8113
INFO:tensorflow:loss = 1.2447264, step = 234340 (7.813 sec)
INFO:tensorflow:lr = 3.7950056e-05 (7.813 sec)
INFO:tensorflow:global_step/sec: 12.5906
INFO:tensorflow:loss = 1.2432296, step = 234440 (7.941 sec)
INFO:tensorflow:lr = 3.791193e-05 (7.942 sec)
INFO:tensorflow:global_step/sec: 12.2831
INFO:tensorflow:loss = 1.2440478, step = 234540 (8.141 sec)
INFO:tensorflow:lr = 3.787385e-05 (8.140 sec)
INFO:tensorflow:global_step/sec: 12.4469
INFO:tensorflow:loss = 1.2417731, step = 234640 (8.028 sec)
INFO:tensorflow:lr = 3.78358e-05 (8.029 sec)
INFO:tensorflow:Saving checkpoints for 234720 into ../model/lstm_transformer/model.ckpt.
INFO:tensorflow:Loss for final step: 1.2419497.
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-234720
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
------------
minimal test
utterance: what times are the nutcracker show playing near me
parsed: [ in:get_event what times are [ sl:category_event the nutcracker show ] playing [ sl:location [ in:get_location [ sl:search_radius near ] [ sl:location_user me ] ] ] ]
in:get_event
__________________________|_____________________________________________________
| | | | | sl:location
| | | | | |
| | | | | in:get_location
| | | | | ________________|_______________
| | | | sl:category_even sl:search_radius sl:location_user
| | | | t | |
| | | | ______________|__________ | |
what times are playing the nutcracker show near me
------------
Reading ../data/test.tsv
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from ../model/lstm_transformer/model.ckpt-234720
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Reading ../data/test.tsv