mnist_dnn.py

Go to the documentation of this file.

'''Trains a simple deep NN on the MNIST dataset.

Gets to 98.40% test accuracy after 20 epochs
(there is *a lot* of margin for parameter tuning).
2 seconds per epoch on a K520 GPU.
'''
# Taken from
# https://raw.githubusercontent.com/keras-team/keras/master/examples/mnist_mlp.py


from __future__ import print_function

import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout, Softmax
from keras.optimizers import RMSprop, SGD, Adam
from keras import regularizers
from keras.constraints import MaxNorm

batch_size = 128
num_classes = 10
epochs = 25

# the data, 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)
l2 = 0.0001
model = Sequential()
model.add(Dense(100, activation='relu', input_shape=(784,),use_bias=False, kernel_regularizer=regularizers.l2(l2)))
#model.add(Dropout(0.3))
model.add(Dense(100, activation='relu', use_bias=False,kernel_regularizer=regularizers.l2(l2)))
#model.add(Dropout(0.4))
model.add(Dense(num_classes, activation='softmax', use_bias=False,kernel_regularizer=regularizers.l2(l2)))

model.summary()

model.compile(#loss='mean_squared_error',
            loss='categorical_crossentropy',
            #optimizer=SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True),
            optimizer=Adam(lr=0.001),
            metrics=['accuracy'])


history = model.fit(x_train, y_train,
                    batch_size=batch_size,
                    epochs=epochs,
                    verbose=1,
                    validation_data=(x_test, y_test))
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])

model.save_weights('dnn_weights.h5')
model.save('dnn_model.h5', include_optimizer=False)

json_string = model.to_json()
with open('dnn_model.json', 'w') as file:
    file.write(json_string)