train_images = train_images.reshape((60000, 28 * 28))
train_images = train_images.astype('float32') / 255
test_images = test_images.reshape((10000, 28 * 28))
test_images = test_images.astype('float32') / 255
# 标签处理
train_labels = to_categorical(train_labels)
test_labels = to_categorical(test_labels)
# 构建模型
model = models.Sequential()
model.add(layers.Dense(512, activation='relu', input_shape=(28*28,)))
model.add(layers.Dropout(0.6))
model.add(layers.Dense(256, activation='relu', input_shape=(28*28,)))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(128, activation='relu', input_shape=(28*28,)))
model.add(layers.Dropout(0.4))
model.add(layers.Dense(10, activation='softmax'))
model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])
time_start = time.time()
history = model.fit(train_images, train_labels, epochs=80, batch_size=128,validation_data=(test_images, test_labels))
print("训练时间：{}s\n".format(int(time.time()-time_start)))
# # 分析结果
test_loss, test_acc = model.evaluate(test_images, test_labels)
print("test_acc:{}\n".format(test_acc))
history_dict = history.history
plot_utils.plot_history(history_dict)
layers.Dense(64,
activation='relu',
kernel_regularizer=regularizers.l2(0.001),
input_shape=(10000, )))
model.add(layers.Dropout(0.6))
model.add(
layers.Dense(32,
activation='relu',
kernel_regularizer=regularizers.l2(0.001)))
model.add(layers.Dropout(0.4))
model.add(layers.Dense(46, activation='softmax'))
model.compile(optimizer=optimizers.Adam(),
loss=losses.categorical_crossentropy,
metrics=['accuracy'])
# 留出验证集
x_val = x_train[:1000]
partial_x_train = x_train[1000:]
y_val = train_labels[:1000]
partial_y_train = train_labels[1000:]
history = model.fit(partial_x_train,
partial_y_train,
epochs=50,
batch_size=512,
validation_data=(x_val, y_val))
results = model.evaluate(x_val, y_val)
print("results:{}\n".format(results))
print("predict:{}\n".format(model.predict(x_val)[20]))
plot_utils.plot_history(history.history)
