Vgg16 migration learning source code

import tensorflow as tf
import numpy as np
import os
from matplotlib import pyplot as plt

Data enhancement

from tensorflow.keras.applications.vgg16 import VGG16
class myNet:
def init(self):
# For training data sets , Data enhancement is used
self.generatorForTrain = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1./255,
rotation_range=20, width_shift_range=0.2,
height_shift_range=0.2, horizontal_flip=True)
# For non training datasets , No data enhancement
self.generatorForText = tf.keras.preprocessing.image.ImageDataGenerator(
rescale=1./255)
self.train=self.generatorForTrain.flow_from_directory(“C:\Users\104005162\PycharmProjects\pythonProject\data\train”,
target_size=(150,150),batch_size=32,class_mode=“binary”,shuffle=True)
self.text=self.generatorForText.flow_from_directory(“C:\Users\104005162\PycharmProjects\pythonProject\data\validation\”,
target_size=(150,150),batch_size=32,class_mode=“binary”,shuffle=True)
self.baseModel=VGG16(include_top=False)

 def getDateSets(self):
     return self.train,self.text
 # Modify the model 
 def refineBaseModel(self):
     x=self.baseModel.output
     x=tf.keras.layers.GlobalAveragePooling2D()(x)
     x=tf.keras.layers.Dense(1024,activation='relu')(x)
     x=tf.keras.layers.Dense(10,activation='softmax')(x)
     self.combineModel=tf.keras.models.Model(inputs=self.baseModel.inputs,outputs=x)

 def freezeModel(self):
     for layer in self.baseModel.layers:
         layer.trainable=False
 def complie(self):
     modelCkpt=tf.keras.callbacks.ModelCheckpoint('./ckpt/singlenn_{epoch:02d}-{val_acc:.2f}.h5',monitor="val_acc",save_weights_only=True,save_best_only=True,period=1)
     self.combineModel.fit_generator(self.train,epochs=3,validation_data=self.text,callbacks=[modelCkpt])

 def fit(self):
     self.combineModel.fit_generator(self.train,epochs=3,validation_data=self.text)

# The following code tests the data set printing of neural network .
dataTool=myNet()
dataTool.getDateSets()
dataTool.freezeModel()
dataTool.refineBaseModel()
dataTool.complie()
dataTool.fit()