10 monkey | CNN | Tensorflow | Kaggle|迁移学习

读取图片和其标签，保存成模型读取的格式

前面代码和上一篇文章一样

# 做卷积的时候所有图片的尺寸应该是一样的
height = 224
width = 224
channels = 3 
batch_size = 24
num_classes = 10

# 读取训练数据并作数据增强
# 确定一些读取格式要求
train_datagen = keras.preprocessing.image.ImageDataGenerator(
    preprocessing_function = keras.applications.resnet50.preprocess_input,
    # 图片旋转的角度范围，用来数据增强
    rotation_range = 40,
    # 水平平移
    width_shift_range = 0.2,
    # 高度平移
    height_shift_range = 0.2,
    # 剪切强度
    shear_range = 0.2,
    # 缩放强度
    zoom_range = 0.2,
    # 水平翻转
    horizontal_flip = True,
    # 对图片做处理时需要填充图片，用最近的像素点填充
    fill_mode = "nearest"
)
# 读取训练数据
train_generator = train_datagen.flow_from_directory(
    train_dir,
    # 读取后将图片存什么大小 
    target_size = (height, width),
    batch_size = batch_size,
    seed = 7,
    shuffle = True,
    # label的编码格式：这里为one-hot编码
    class_mode = 'categorical')

# 读取验证数据
valid_datagen = keras.preprocessing.image.ImageDataGenerator(preprocessing_function = keras.applications.resnet50.preprocess_input)
valid_generator = valid_datagen.flow_from_directory(
    valid_dir,
    # 读取后将图片存什么大小 
    target_size = (height, width),
    batch_size = batch_size,
    seed = 7,
    shuffle = False,
    # label的编码格式：这里为one-hot编码
    class_mode = 'categorical')

train_num = train_generator.samples
valid_num = valid_generator.samples
print(train_num, valid_num)

模型定义

仅训练ResRet最后一层的模型

from keras.applications.resnet50 import ResNet50
resnet50_fine_tune = keras.models.Sequential()
# resnet50最后一层有1000类，所以在本例中把最后一层去掉（include_top = False）
# resnet的倒数第二层是一个三维矩阵，所以无法与全连接层连接，故要pooling
# weights = 'imagenet'： 下载一个模型，然后在这个模型的基础上进行训练
# weights = 'None'：从头开始训练
resnet50_fine_tune.add(ResNet50(include_top = False,
                                pooling = 'avg',
                                weights = 'imagenet'))
resnet50_fine_tune.add(keras.layers.Dense(num_classes,activation = 'softmax'))
resnet50_fine_tune.layer[0].trainable = False

resnet50_fine_tune.compile(loss="categorical_crossentropy",
             optimizer="sgd",
             metrics = ["accuracy"])
resnet50_fine_tune.summary()

可训练参数为两千多

训练最后几层的模型

resnet50 = keras.applications.ResNet50(include_top=False, pooling = 'avg', weights='imagenet')
resnet50.summary()

# 将后五层设置为可训练
for layer in resnet50.layers[0:-5]:
    layer.trainable = False

resnet50_new = keras.models.Sequential([resnet50, keras.layers.Dense(num_classes, activation = 'softmax')])

resnet50_new.compile(loss="categorical_crossentropy",
             optimizer="sgd",
             metrics = ["accuracy"])
resnet50_new.summary()

可训练参数为一百万多

训练网络

epochs = 10 
# 数据是generator出来的，所以不能直接用fit
history = resnet50_new.fit_generator(train_generator,
                                     steps_per_epoch = train_num // batch_size,
                                     epochs=epochs,
                                     validation_data = valid_generator,
                                     validation_steps = valid_num // batch_size)