《Tensorflow + Keras深度学习人工智能实践应用》 一书第8章的完整例子，相对于第7章的感知机，第八章使用了神经网络，进一步提高了准确率。
在实现上，主要增加了卷积层和池化层。卷积层，使用了filters（即使用多少个卷积核）和卷积核的大小，以及做卷积的填充方式。池化，主要缩减了图片的大小。

import numpy as np
import pandas as pd
from keras.utils import np_utils
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import Dropout, Flatten, Conv2D, MaxPooling2D

import matplotlib.pyplot as plt
import os

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
np.random.seed(10)

(x_train_image, y_train_label), (x_test_image, y_test_label) = mnist.load_data()

x_train = x_train_image.reshape(x_train_image.shape[0], 28, 28, 1).astype("float32")
x_test = x_test_image.reshape(x_test_image.shape[0], 28, 28, 1).astype("float32")

x_train_normal = x_train / 255
x_test_normal = x_test / 255

y_train_onehot = np_utils.to_categorical(y_train_label)
y_test_onehot = np_utils.to_categorical(y_test_label)

model = Sequential()

model.add(Conv2D(filters=16,
                kernel_size = (5, 5),
                padding = 'same',
                input_shape = (28, 28, 1),
                activation = 'relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(filters=36,
                kernel_size = (5, 5),
                padding = 'same',
                activation = 'relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))

model.add(Flatten())
model.add(Dense(units = 128,
                activation = 'relu'))
model.add(Dropout(0.5))
model.add(Dense(units = 10,
                kernel_initializer = 'normal',
                activation = 'softmax'))

print(model.summary())


model.compile(loss = "categorical_crossentropy",
            optimizer = "adam", metrics = ["accuracy"])

history = model.fit(x = x_train_normal,
                y = y_train_onehot,
                validation_split = 0.2,
                epochs = 10,
                batch_size = 300,
                verbose = 2)

def show_train_history(train_history, train, val):
    plt.plot(train_history.history[train])
    plt.plot(train_history.history[val])
    plt.title("Train History")
    plt.ylabel(train)
    plt.xlabel("Epochs")
    plt.legend(["train", "validation"], loc="upper left")
    plt.show()

def plot_image_label_prediction(images, labels, prediction, idx = 0, num = 10):
    fig = plt.gcf()
    fig.set_size_inches(12, 14)
    if num > 25:
        num = 25
    for i in range(0, num):
        ax = plt.subplot(5, 5, 1 + i)
        ax.imshow(images[idx], cmap="binary")
        title = "label = " + str(labels[idx])
        if len(prediction) > 0:
            title += ", prediction = " + str(prediction[idx])
        ax.set_title(title, fontsize = 12)
        ax.set_xticks([])
        ax.set_yticks([])
        idx += 1
    plt.show()


show_train_history(history, "acc", "val_acc")
show_train_history(history, "loss", "val_loss")

scores = model.evaluate(x_test_normal, y_test_onehot)
print("accuracy = ", scores[1])

prediction = model.predict_classes(x_test_normal)
#plot_image_label_prediction(x_test_image, y_test_label, prediction, idx=340, num=25)

print(pd.crosstab(y_test_label, prediction, rownames = ["label"], colnames = ["predict"]))

df = pd.DataFrame({"label": y_test_label, "predict": prediction})
print(df[(df.label == 5) & (df.predict == 3)])

训练及精度：

Train on 48000 samples, validate on 12000 samples
Epoch 1/10
 - 43s - loss: 0.5479 - acc: 0.8286 - val_loss: 0.1138 - val_acc: 0.9670
Epoch 2/10
 - 44s - loss: 0.1516 - acc: 0.9541 - val_loss: 0.0766 - val_acc: 0.9765
Epoch 3/10
 - 46s - loss: 0.1127 - acc: 0.9663 - val_loss: 0.0575 - val_acc: 0.9827
Epoch 4/10
 - 46s - loss: 0.0937 - acc: 0.9721 - val_loss: 0.0502 - val_acc: 0.9857
Epoch 5/10
 - 46s - loss: 0.0812 - acc: 0.9756 - val_loss: 0.0457 - val_acc: 0.9870
Epoch 6/10
 - 45s - loss: 0.0693 - acc: 0.9789 - val_loss: 0.0426 - val_acc: 0.9879
Epoch 7/10
 - 46s - loss: 0.0617 - acc: 0.9808 - val_loss: 0.0410 - val_acc: 0.9883
Epoch 8/10
 - 44s - loss: 0.0568 - acc: 0.9830 - val_loss: 0.0375 - val_acc: 0.9896
Epoch 9/10
 - 44s - loss: 0.0523 - acc: 0.9845 - val_loss: 0.0358 - val_acc: 0.9898
Epoch 10/10
 - 43s - loss: 0.0468 - acc: 0.9859 - val_loss: 0.0360 - val_acc: 0.9895
10000/10000 [==============================] - 4s 389us/step
accuracy =  0.9922