Multi-label DistilBert model trained (/fine-tuned) on toxicity dataset:
# !conda install watermark -c conda-forge --yes
%load_ext watermark
%watermark -p torch,transformers,pandas,tqdm
torch : 1.11.0 transformers: 4.20.1 pandas : 1.3.5 tqdm : 4.64.0
import pandas as pd
import torch
from tqdm import tqdm
from torch.utils.data import Dataset, DataLoader, RandomSampler, SequentialSampler
from transformers import DistilBertTokenizer, DistilBertModel
MAX_LEN = 512
TRAIN_BATCH_SIZE = 16
VALID_BATCH_SIZE = 16
EPOCHS = 3
LEARNING_RATE = 1e-05
DEVICE = 'cuda:0' if torch.cuda.is_available() else 'cpu'
print(DEVICE)
cuda:0
train_data = pd.read_csv('/kaggle/input/jigsaw-toxic-comment-classification-challenge/train.csv.zip')
label_columns = ["toxic", "severe_toxic", "obscene", "threat", "insult", "identity_hate"]
train_data['labels'] = train_data[label_columns].apply(lambda x: list(x), axis=1)
train_data.drop(['id'], inplace=True, axis=1)
train_data.drop(label_columns, inplace=True, axis=1)
train_data.head()
| comment_text | labels | |
|---|---|---|
| 0 | Explanation\nWhy the edits made under my usern... | [0, 0, 0, 0, 0, 0] |
| 1 | D'aww! He matches this background colour I'm s... | [0, 0, 0, 0, 0, 0] |
| 2 | Hey man, I'm really not trying to edit war. It... | [0, 0, 0, 0, 0, 0] |
| 3 | "\nMore\nI can't make any real suggestions on ... | [0, 0, 0, 0, 0, 0] |
| 4 | You, sir, are my hero. Any chance you remember... | [0, 0, 0, 0, 0, 0] |
class MultiLabelDataset(Dataset):
def __init__(self, dataframe, tokenizer, max_len, new_data=False):
self.tokenizer = tokenizer
self.data = dataframe
self.text = dataframe.comment_text
self.new_data = new_data
if not new_data:
self.targets = self.data.labels
self.max_len = max_len
def __len__(self):
return len(self.text)
def __getitem__(self, index):
text = str(self.text[index])
text = " ".join(text.split())
inputs = self.tokenizer.encode_plus(
text,
None,
add_special_tokens=True,
max_length=self.max_len,
pad_to_max_length=True,
return_token_type_ids=True
)
ids = inputs['input_ids']
mask = inputs['attention_mask']
token_type_ids = inputs["token_type_ids"]
out = {
'ids': torch.tensor(ids, dtype=torch.long),
'mask': torch.tensor(mask, dtype=torch.long),
'token_type_ids': torch.tensor(token_type_ids, dtype=torch.long),
}
if not self.new_data:
out['targets'] = torch.tensor(self.targets[index], dtype=torch.float)
return out
train_size = 1.0
train_df = train_data.sample(frac=train_size, random_state=123)
val_df = train_data.drop(train_df.index).reset_index(drop=True)
train_df = train_df.reset_index(drop=True)
print("Orig Dataset: {}".format(train_data.shape))
print("Training Dataset: {}".format(train_df.shape))
print("Validation Dataset: {}".format(val_df.shape))
tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased', truncation=True, do_lower_case=True)
training_set = MultiLabelDataset(train_df, tokenizer, MAX_LEN)
val_set = MultiLabelDataset(val_df, tokenizer, MAX_LEN)
Orig Dataset: (159571, 2) Training Dataset: (159571, 2) Validation Dataset: (0, 2)
Downloading: 0%| | 0.00/226k [00:00<?, ?B/s]
Downloading: 0%| | 0.00/28.0 [00:00<?, ?B/s]
Downloading: 0%| | 0.00/483 [00:00<?, ?B/s]
train_params = {'batch_size': TRAIN_BATCH_SIZE,
'shuffle': True,
'num_workers': 8
}
val_params = {'batch_size': VALID_BATCH_SIZE,
'shuffle': False,
'num_workers': 8
}
training_loader = DataLoader(training_set, **train_params)
#val_loader = DataLoader(val_set, **val_params)
/opt/conda/lib/python3.7/site-packages/torch/utils/data/dataloader.py:490: UserWarning: This DataLoader will create 8 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary. cpuset_checked))
class DistilBERTClass(torch.nn.Module):
def __init__(self):
super(DistilBERTClass, self).__init__()
self.bert = DistilBertModel.from_pretrained("distilbert-base-uncased")
self.classifier = torch.nn.Sequential(
torch.nn.Linear(768, 768),
torch.nn.ReLU(),
torch.nn.Dropout(0.1),
torch.nn.Linear(768, 6)
)
def forward(self, input_ids, attention_mask, token_type_ids):
output_1 = self.bert(input_ids=input_ids, attention_mask=attention_mask)
hidden_state = output_1[0]
out = hidden_state[:, 0]
out = self.classifier(out)
return out
model = DistilBERTClass()
model.to(DEVICE);
Downloading: 0%| | 0.00/256M [00:00<?, ?B/s]
Some weights of the model checkpoint at distilbert-base-uncased were not used when initializing DistilBertModel: ['vocab_transform.bias', 'vocab_layer_norm.bias', 'vocab_layer_norm.weight', 'vocab_transform.weight', 'vocab_projector.bias', 'vocab_projector.weight'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
optimizer = torch.optim.Adam(params=model.parameters(), lr=LEARNING_RATE)
def train(epoch):
model.train()
for _, data in tqdm(enumerate(training_loader, 0)):
ids = data['ids'].to(DEVICE, dtype=torch.long)
mask = data['mask'].to(DEVICE, dtype=torch.long)
token_type_ids = data['token_type_ids'].to(DEVICE, dtype=torch.long)
targets = data['targets'].to(DEVICE, dtype=torch.float)
outputs = model(ids, mask, token_type_ids)
optimizer.zero_grad()
loss = torch.nn.functional.binary_cross_entropy_with_logits(outputs, targets)
if _ % 5000 == 0:
print(f'Epoch: {epoch}, Loss: {loss.item()}')
loss.backward()
optimizer.step()
for epoch in range(EPOCHS):
train(epoch)
Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. 0it [00:00, ?it/s]Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning,
Epoch: 0, Loss: 0.7045817375183105
5000it [34:46, 2.41it/s]
Epoch: 0, Loss: 0.0034446530044078827
9974it [1:09:20, 2.40it/s] Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, 0it [00:00, ?it/s]Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning,
Epoch: 1, Loss: 0.08794672787189484
5000it [34:50, 2.40it/s]
Epoch: 1, Loss: 0.01744117960333824
9974it [1:09:32, 2.39it/s] Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. 0it [00:00, ?it/s]Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning,
Epoch: 2, Loss: 0.024290692061185837
5000it [34:53, 2.40it/s]
Epoch: 2, Loss: 0.008041102439165115
9974it [1:09:34, 2.39it/s]
test_data = pd.read_csv('/kaggle/input/jigsaw-toxic-comment-classification-challenge/test.csv.zip')
test_data.head()
| id | comment_text | |
|---|---|---|
| 0 | 00001cee341fdb12 | Yo bitch Ja Rule is more succesful then you'll... |
| 1 | 0000247867823ef7 | == From RfC == \n\n The title is fine as it is... |
| 2 | 00013b17ad220c46 | " \n\n == Sources == \n\n * Zawe Ashton on Lap... |
| 3 | 00017563c3f7919a | :If you have a look back at the source, the in... |
| 4 | 00017695ad8997eb | I don't anonymously edit articles at all. |
test_set = MultiLabelDataset(test_data, tokenizer, MAX_LEN, new_data=True)
test_loader = DataLoader(test_set, **val_params)
/opt/conda/lib/python3.7/site-packages/torch/utils/data/dataloader.py:490: UserWarning: This DataLoader will create 8 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary. cpuset_checked))
all_test_pred = []
def test(epoch):
model.eval()
with torch.inference_mode():
for _, data in tqdm(enumerate(test_loader, 0)):
ids = data['ids'].to(DEVICE, dtype=torch.long)
mask = data['mask'].to(DEVICE, dtype=torch.long)
token_type_ids = data['token_type_ids'].to(DEVICE, dtype=torch.long)
outputs = model(ids, mask, token_type_ids)
probas = torch.sigmoid(outputs)
all_test_pred.append(probas)
return probas
probas = test(model)
Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. 0it [00:00, ?it/s]/opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`. /opt/conda/lib/python3.7/site-packages/transformers/tokenization_utils_base.py:2307: FutureWarning: The `pad_to_max_length` argument is deprecated and will be removed in a future version, use `padding=True` or `padding='longest'` to pad to the longest sequence in the batch, or use `padding='max_length'` to pad to a max length. In this case, you can give a specific length with `max_length` (e.g. `max_length=45`) or leave max_length to None to pad to the maximal input size of the model (e.g. 512 for Bert). FutureWarning, 9573it [21:20, 7.48it/s]
all_test_pred = torch.cat(all_test_pred)
submit_df = test_data.copy()
submit_df.drop("comment_text", inplace=True, axis=1)
label_columns = ["toxic", "severe_toxic", "obscene", "threat", "insult", "identity_hate"]
for i,name in enumerate(label_columns):
submit_df[name] = all_test_pred[:, i].cpu()
submit_df.head()
submit_df.to_csv('submission.csv', index=False)