VGG16实现MNIST分类

# -*- coding = utf-8 -*-
# @Time : 2021/7/26
# @Author : pistachio
# @File : P29.py
# @Software : PyCharm

from keras.models import Sequential
from keras.layers import Conv2D, Activation, MaxPooling2D, Flatten, Dense, Dropout, Reshape
from keras.optimizers import RMSprop
from keras.utils.vis_utils import plot_model
from keras.datasets import mnist
from keras.utils import np_utils
import os
os.environ['PATH'] += os.pathsep + r'C:\Program Files\Graphviz\bin'
import matplotlib.pyplot as plt

# preprocess mnist dataset
(X_train, Y_train), (X_test, Y_test) = mnist.load_data()
X_test1 = X_test
Y_test1 = Y_test
X_train = X_train.reshape(-1, 28, 28, 1)
X_test = X_test.reshape(-1, 28, 28, 1)
X_train = X_train.reshape(X_train.shape[0], 28, 28, 1)
X_test = X_test.reshape(X_test.shape[0], 28, 28, 1)
X_train = X_train.astype('float32') / 255.0
X_test = X_test.astype('float32') / 255.0
Y_train = np_utils.to_categorical(Y_train, 10)
Y_test = np_utils.to_categorical(Y_test, 10)

# use Sequential model build VGG16 network structure
def vgg16():
    model = Sequential()
    model.add(Reshape((28, 28, 1)))
    model.add(Conv2D(64, (3, 3), padding='same',
                    input_shape=((28, 28, 1)), activation='relu'))
    model.add(Conv2D(64, (3, 3), padding='same', activation='relu'))
    
    model.add(MaxPooling2D(data_format='channels_last', pool_size=(2, 2)))
    
    model.add(Conv2D(128, (3, 3), padding='same', activation='relu'))
    model.add(Conv2D(128, (3, 3), padding='same', activation='relu'))
    
    model.add(MaxPooling2D(data_format='channels_last', pool_size=(2, 2)))
    
    model.add(Conv2D(256, (3, 3), padding='same', activation='relu'))
    model.add(Conv2D(256, (3, 3), padding='same', activation='relu'))
    model.add(Conv2D(256, (3, 3), padding='same', activation='relu'))
    
    model.add(MaxPooling2D(data_format='channels_last', pool_size=(2, 2)))
    
    model.add(Conv2D(512, (3, 3), padding='same', activation='relu'))
    model.add(Conv2D(512, (3, 3), padding='same', activation='relu'))
    model.add(Conv2D(512, (3, 3), padding='same', activation='relu'))
    
    model.add(MaxPooling2D(data_format='channels_last', pool_size=(2, 2)))
    
    model.add(Conv2D(512, (3, 3), padding='same', activation='relu'))
    model.add(Conv2D(512, (3, 3), padding='same', activation='relu'))
    model.add(Conv2D(512, (3, 3), padding='same', activation='relu'))
    
    model.add(Flatten())
    
    model.add(Dense(256, activation='relu'))
    model.add(Dropout(0.5))
    
    model.add(Dense(256, activation='relu'))
    model.add(Dropout(0.5))
    
    model.add(Dense(10))
    model.add(Activation('softmax'))
    
    model.compile(
        loss='binary_crossentropy',
        optimizer=RMSprop(lr=1e-4),
        metrics=['accuracy'])
    return model

model = vgg16()
model.build(input_shape=( 100, 28, 28, 1))
model.summary()

# the compiled vgg16 network structure is trained and predicted
np_epoch = 4
batch_size = 256


def run_vgg16():
    Training = model.fit(X_train, Y_train, batch_size=batch_size,
                         epochs=np_epoch, validation_split=0.25, verbose=1)
    predictions = model.predict(X_test, verbose=0)
    return predictions, Training
predictions, Training = run_vgg16()

# visual vgg16 network structure
plot_model(model=model, to_file='model_VGG16.png', show_shapes=True)

# evaluate and predict VGG16 network model
# use test set evaluate vgg16 network model's accuracy
Test = model.evaluate(X_test, Y_test)
print('loss=', Test[0])
print('accuracy', Test[1])

# draw accuracy curve an val_acc curve during vgg16 training
def show_Training_acc_history(Training_history, train, validation):
    plt.plot(Training.history[train], linestyle='-', color='b')
    plt.plot(Training.history[validation], linestyle='--', color='r')
    plt.title('Training accuracy history')
    plt.xlabel('epoch')
    plt.ylabel('train')
    plt.legend(['train', 'validation'], loc='upper right')
    plt.show()
show_Training_acc_history(Training, 'acc', 'val_acc')

def show_Training_loss_history(Training_history, train, validation):
    plt.plot(Training.history[train], linestyle='-', color='b')
    plt.plot(Training.history[validation], linestyle='--', color='r')
    plt.title('Training loss history')
    plt.xlabel('epoch')
    plt.ylabel('train')
    plt.legend(['train', 'validation'], loc='lower right')
    plt.show()
show_Training_loss_history(Training, 'loss', 'val_loss')

# use vgg16 network to make predictions
def plot_image(image):
    fig = plt.gcf()
    fig.set_size_inches(2, 2)
    plt.imshow(image, cmap='binary')
    plt.show()

def pre_results(i):
    plot_image(X_test1[i])
    print('Y_test1=', Y_test1[i])
    print('pre_result=', predictions[i])
pre_results(0)