In [1]:
from __future__ import print_function
import tensorflow as tf
import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras import backend as K
In [2]:
batch_size = 128
num_classes = 10
epochs = 12
# input image dimensions
img_rows, img_cols = 28, 28
In [3]:
# the data, split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()
if K.image_data_format() == 'channels_first':
x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
input_shape = (1, img_rows, img_cols)
else:
x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)
x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)
input_shape = (img_rows, img_cols, 1)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
print('x_train shape:', x_train.shape)
print(x_train.shape[0], 'train samples')
print(x_test.shape[0], 'test samples')
# convert class vectors to binary class matrices
y_train = keras.utils.np_utils.to_categorical(y_train, num_classes)
y_test = keras.utils.np_utils.to_categorical(y_test, num_classes)
x_train shape: (60000, 28, 28, 1) 60000 train samples 10000 test samples
Model description
In [4]:
model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3),
activation='relu',
input_shape=input_shape))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes, activation='softmax'))
Visualize the model
In [5]:
from keras.utils.vis_utils import plot_model
plot_model(model, show_shapes=True, show_layer_names=True)
Out[5]:
In [6]:
model.compile(loss=keras.losses.categorical_crossentropy,
optimizer=tf.keras.optimizers.Adadelta(),
metrics=['accuracy'])
In [7]:
model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_data=(x_test, y_test))
Epoch 1/12 469/469 [==============================] - 152s 322ms/step - loss: 2.2853 - accuracy: 0.1440 - val_loss: 2.2401 - val_accuracy: 0.3025 Epoch 2/12 469/469 [==============================] - 149s 319ms/step - loss: 2.2096 - accuracy: 0.2709 - val_loss: 2.1427 - val_accuracy: 0.5774 Epoch 3/12 469/469 [==============================] - 152s 324ms/step - loss: 2.1070 - accuracy: 0.3898 - val_loss: 2.0075 - val_accuracy: 0.6674 Epoch 4/12 469/469 [==============================] - 150s 320ms/step - loss: 1.9678 - accuracy: 0.4766 - val_loss: 1.8295 - val_accuracy: 0.7110 Epoch 5/12 469/469 [==============================] - 151s 321ms/step - loss: 1.7959 - accuracy: 0.5395 - val_loss: 1.6154 - val_accuracy: 0.7498 Epoch 6/12 469/469 [==============================] - 149s 319ms/step - loss: 1.6022 - accuracy: 0.5897 - val_loss: 1.3878 - val_accuracy: 0.7750 Epoch 7/12 469/469 [==============================] - 150s 319ms/step - loss: 1.4190 - accuracy: 0.6272 - val_loss: 1.1815 - val_accuracy: 0.7942 Epoch 8/12 469/469 [==============================] - 148s 317ms/step - loss: 1.2640 - accuracy: 0.6551 - val_loss: 1.0142 - val_accuracy: 0.8068 Epoch 9/12 469/469 [==============================] - 149s 318ms/step - loss: 1.1364 - accuracy: 0.6782 - val_loss: 0.8848 - val_accuracy: 0.8194 Epoch 10/12 469/469 [==============================] - 150s 321ms/step - loss: 1.0415 - accuracy: 0.6967 - val_loss: 0.7887 - val_accuracy: 0.8289 Epoch 11/12 469/469 [==============================] - 148s 315ms/step - loss: 0.9647 - accuracy: 0.7142 - val_loss: 0.7151 - val_accuracy: 0.8369 Epoch 12/12 469/469 [==============================] - 148s 315ms/step - loss: 0.9049 - accuracy: 0.7269 - val_loss: 0.6575 - val_accuracy: 0.8446
Out[7]:
<keras.callbacks.History at 0x7f20df231810>
In [8]:
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
Test loss: 0.6575264930725098 Test accuracy: 0.8446000218391418