In [6]:

    

import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout
from keras.optimizers import RMSprop
import keras

batch_size = 2048
num_classes = 10
epochs = 20

# the data, shuffled and split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()

x_train = x_train.reshape(60000, 784)
x_test = x_test.reshape(10000, 784)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
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.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)

model = Sequential()
model.add(Dense(512, activation='relu', input_shape=(784,)))
model.add(Dropout(0.2))
model.add(Dense(512, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))

model.summary()

model.compile(loss='categorical_crossentropy',
              optimizer=RMSprop(),
              metrics=['accuracy'])

tb = keras.callbacks.TensorBoard(log_dir='./logs/2')

history = model.fit(x_train, y_train,
                    batch_size=batch_size,
                    epochs=epochs,
                    verbose=1,
                    validation_data=(x_test, y_test),
                    callbacks=[tb])

score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])


    

    




60000 train samples
10000 test samples
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_10 (Dense)             (None, 512)               401920    
_________________________________________________________________
dropout_7 (Dropout)          (None, 512)               0         
_________________________________________________________________
dense_11 (Dense)             (None, 512)               262656    
_________________________________________________________________
dropout_8 (Dropout)          (None, 512)               0         
_________________________________________________________________
dense_12 (Dense)             (None, 10)                5130      
=================================================================
Total params: 669,706
Trainable params: 669,706
Non-trainable params: 0
_________________________________________________________________
Train on 60000 samples, validate on 10000 samples
Epoch 1/20
60000/60000 [==============================] - 8s - loss: 0.7039 - acc: 0.7822 - val_loss: 0.3642 - val_acc: 0.8867
Epoch 2/20
60000/60000 [==============================] - 6s - loss: 0.2873 - acc: 0.9137 - val_loss: 0.1910 - val_acc: 0.9412
Epoch 3/20
60000/60000 [==============================] - 7s - loss: 0.2050 - acc: 0.9381 - val_loss: 0.1394 - val_acc: 0.9581
Epoch 4/20
60000/60000 [==============================] - 7s - loss: 0.1621 - acc: 0.9504 - val_loss: 0.1207 - val_acc: 0.9643
Epoch 5/20
60000/60000 [==============================] - 7s - loss: 0.1293 - acc: 0.9607 - val_loss: 0.1089 - val_acc: 0.9666
Epoch 6/20
60000/60000 [==============================] - 8s - loss: 0.1080 - acc: 0.9667 - val_loss: 0.1329 - val_acc: 0.9567
Epoch 7/20
60000/60000 [==============================] - 8s - loss: 0.0884 - acc: 0.9727 - val_loss: 0.0913 - val_acc: 0.9711
Epoch 8/20
60000/60000 [==============================] - 9s - loss: 0.0770 - acc: 0.9754 - val_loss: 0.0827 - val_acc: 0.9737
Epoch 9/20
60000/60000 [==============================] - 7s - loss: 0.0654 - acc: 0.9799 - val_loss: 0.0760 - val_acc: 0.9749
Epoch 10/20
60000/60000 [==============================] - 8s - loss: 0.0584 - acc: 0.9817 - val_loss: 0.0643 - val_acc: 0.9804
Epoch 11/20
60000/60000 [==============================] - 8s - loss: 0.0485 - acc: 0.9844 - val_loss: 0.0616 - val_acc: 0.9812
Epoch 12/20
60000/60000 [==============================] - 8s - loss: 0.0441 - acc: 0.9860 - val_loss: 0.0639 - val_acc: 0.9805
Epoch 13/20
60000/60000 [==============================] - 8s - loss: 0.0383 - acc: 0.9877 - val_loss: 0.0637 - val_acc: 0.9795
Epoch 14/20
60000/60000 [==============================] - 8s - loss: 0.0336 - acc: 0.9890 - val_loss: 0.0609 - val_acc: 0.9813
Epoch 15/20
60000/60000 [==============================] - 7s - loss: 0.0282 - acc: 0.9911 - val_loss: 0.0612 - val_acc: 0.9817
Epoch 16/20
60000/60000 [==============================] - 8s - loss: 0.0242 - acc: 0.9922 - val_loss: 0.0788 - val_acc: 0.9771
Epoch 17/20
60000/60000 [==============================] - 6s - loss: 0.0238 - acc: 0.9927 - val_loss: 0.0583 - val_acc: 0.9838
Epoch 18/20
60000/60000 [==============================] - 7s - loss: 0.0236 - acc: 0.9925 - val_loss: 0.0612 - val_acc: 0.9828
Epoch 19/20
60000/60000 [==============================] - 8s - loss: 0.0183 - acc: 0.9942 - val_loss: 0.0676 - val_acc: 0.9813
Epoch 20/20
60000/60000 [==============================] - 8s - loss: 0.0163 - acc: 0.9946 - val_loss: 0.0734 - val_acc: 0.9790
Test loss: 0.0733699267333
Test accuracy: 0.979

In [ ]: