猫狗识别基于tensorflow2.0 GPU版 自建CNN模型+数据增强+Dropout

猫狗识别基于tensorflow2.0 GPU版 自建CNN模型+数据增强+Dropout

1. 导入库

from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
from tensorflow.keras.models import Sequential, load_model
from tensorflow.keras import optimizers
from tensorflow.keras.preprocessing.image import ImageDataGenerator

import matplotlib.pyplot as plt

2. 配置GPU

from tensorflow.compat.v1 import ConfigProto
from tensorflow.compat.v1 import InteractiveSession
# 定义TensorFlow配置
config = ConfigProto()
# 配置GPU内存分配方式，按需增长，很关键
config.gpu_options.allow_growth = True
# 在创建session的时候把config作为参数传进去
session = InteractiveSession(config=config)

3. 图像数据加载和增强

#手动定义数据集目录
train_dir='./catdogdata/train'
validation_dir='./catdogdata/validation'
test_dir='./catdogdata/test'

train_datagen = ImageDataGenerator(rescale=1/255, 
                                   rotation_range=40, 
                                   width_shift_range=0.2, 
                                   height_shift_range=0.2,
                                   shear_range=0.2, 
                                   zoom_range=0.2, 
                                   horizontal_flip=True)
train_generator = train_datagen.flow_from_directory(train_dir