CNNTest.py

import numpy as np
np.random.seed(1337)
from keras.datasets import mnist
from keras.utils import np_utils
from keras.models import Sequential
from keras.layers import Dense, Activation, Convolution2D, MaxPool2D, Flatten
from keras.optimizers import Adam

(X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(-1, 1, 28, 28)
X_test = X_test.reshape(-1, 1, 28, 28)
y_test = np_utils.to_categorical(y_test, num_classes=10)
y_train = np_utils.to_categorical(y_train, num_classes=10)

model = Sequential()
model.add(Convolution2D(
    filters=32, kernel_size=(5, 5), padding='same', input_shape=(1, 28, 28)
))
model.add(Activation('relu'))
model.add(MaxPool2D(
    pool_size=(2, 2), padding='same'
))
model.add(Convolution2D(
    filters=64, kernel_size=(5, 5), padding='same'
))
model.add(MaxPool2D(
    strides=(2, 2), padding='same'
))
model.add(Flatten())
model.add(Dense(1024))
model.add(Activation('relu'))
model.add(Dense(10))
model.add(Activation('softmax'))
adam = Adam()
model.compile(optimizer=adam, loss='categorical_crossentropy', metrics=['accuracy'])
print("Train...")
model.fit(X_train, y_train, epochs=2, batch_size=32, )
print("Test...")
loss, accuracy = model.evaluate(X_test, y_test)
print("loss : ", loss)
print("accuracy : ", accuracy)