vgg16 清华镜像_经典网络-使用VGG16进行迁移学习
# 使用迁移学习的思想，以VGG16作为模板搭建模型，训练识别手写字体
# 引入VGG16模块
from keras.applications.vgg16 import VGG16
# 其次加载其他模块
from keras.layers import Input
from keras.layers import Flatten
from keras.layers import Dense
from keras.layers import Dropout
from keras.models import Model
from keras.optimizers import SGD
# 加载字体库作为训练样本
from keras.datasets import mnist
# 加载OpenCV(在命令行中窗口中输入pip install opencv-python)，这里为了后期对图像的处理，
# 大家使用pip install C:\Users\28542\Downloads\opencv_python-3.4.1+contrib-cp36-cp36m-win_amd64.whl
# 比如尺寸变化和Channel变化。这些变化是为了使图像满足VGG16所需要的输入格式
import cv2
import h5py as h5py
import numpy as np
# 建立一个模型，其类型是Keras的Model类对象，我们构建的模型会将VGG16顶层去掉，只保留其余的网络
# 结构。这里用include_top = False表明我们迁移除顶层以外的其余网络结构到自己的模型中
# VGG模型对于输入图像数据要求高宽至少为48个像素点，由于硬件配置限制，我们选用48个像素点而不是原来
# VGG16所采用的224个像素点。即使这样仍然需要24GB以上的内存，或者使用数据生成器
model_vgg = VGG16(include_top=False, weights='imagenet', input_shape=(48, 48, 3))
for layer in model_vgg.layers:
layer.trainable = False
model = Flatten(name='flatten')(model_vgg.output)
model = Dense(4096, activation='relu', name='fc1')(model)
model = Dense(4096, activation='relu', name='fc2')(model)
model = Dropout(0.5)(model)
model = Dense(10, activation='softmax')(model)
model_vgg_mnist = Model(inputs=model_vgg.input, outputs=model, name='vgg16')
# 打印模型结构，包括所需要的参数
model_vgg_mnist.summary()
'''model_vgg = VGG16(include_top=False, weights='imagenet', input_shape=(224, 224, 3))for layer in model_vgg.layers:layer.trainable = Falsemodel = Flatten()(model_vgg.output)model = Dense(4096, activation='relu', name='fc1')(model)model = Dense(4096, activation='relu', name='fc2')(model)model = Dropout(0.5)(model)model = Dense(10, activation='softmax', name='prediction')(model)model_vgg_mnist_pretrain = Model(model_vgg.input, model, name='vgg16_pretrain')model_vgg_mnist_pretrain.summary()'''
# 新的模型不需要训练原有卷积结构里面的1471万个参数，但是注意参数还是来自于最后输出层前的两个
# 全连接层，一共有1.2亿个参数需要训练
sgd = SGD(lr=0.05, decay=1e-5)
model_vgg_mnist.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['accuracy'])
# 因为VGG16对网络输入层的要求，我们用OpenCV把图像从32*32变成224*224，把黑白图像转成RGB图像
# 并把训练数据转化成张量形式，供keras输入
(X_train, y_train), (X_test, y_test) = mnist.load_data("../test_data_home")
# X_train, y_train = X_train[:10000], y_train[:10000]
# X_test, y_test = X_test[:1000], y_test[:1000]
X_train = [cv2.cvtColor(cv2.resize(i, (48, 48)), cv2.COLOR_GRAY2RGB)
for i in X_train]
# 下面concatenate做的事情是把每个样本按照行堆叠在一起，因为是np下面的方法，所以返回的是ndarray
# np.newaxis它本质是None，arr是(48,48,3)，arr[None]是(1,48,48,3)
X_train = np.concatenate([arr[np.newaxis] for arr in X_train]).astype('float32')
X_test = [cv2.cvtColor(cv2.resize(i, (48, 48)), cv2.COLOR_GRAY2RGB)
for i in X_test]
X_test = np.concatenate([arr[np.newaxis] for arr in X_test]).astype('float32')
print(X_train.shape)
print(X_test.shape)
X_train /= 255
X_test /= 255
def tran_y(y):
y_ohe = np.zeros(10)
y_ohe[y] = 1
return y_ohe
y_train_ohe = np.array([tran_y(y_train[i]) for i in range(len(y_train))])
y_test_ohe = np.array([tran_y(y_test[i]) for i in range(len(y_test))])
model_vgg_mnist.fit(X_train, y_train_ohe, validation_data=(X_test, y_test_ohe),
epochs=100, batch_size=100)
# model_vgg_mnist_pretrain.summary()