python代码实现resnet猫狗二分类源码

ResNet（残差网络）是一种深度卷积神经网络模型，在图像识别任务中非常成功，特别是在ImageNet数据集上。以下是使用Python和Keras库实现ResNet基本结构的一个简化的二分类猫狗分类的示例：

from keras.applications.resnet50 import ResNet50, preprocess_input
from keras.layers import Dense, GlobalAveragePooling2D
from keras.models import Model
from keras.preprocessing.image import ImageDataGenerator

# 加载预训练的ResNet50模型（去掉顶部全连接层）
base_model = ResNet50(weights='imagenet', include_top=False, input_shape=(224, 224, 3))

# 添加全局平均池化和新的全连接层
x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dense(1024, activation='relu')(x)  # 可调整此层的隐藏单元数
predictions = Dense(1, activation='sigmoid')(x)

# 创建新模型
model = Model(inputs=base_model.input, outputs=predictions)

# 冻结基础模型的权重
for layer in base_model.layers:
    layer.trainable = False

# 编译模型
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

# 数据增强
train_datagen = ImageDataGenerator(preprocessing_function=preprocess_input, rotation_range=20, width_shift_range=0.1, height_shift_range=0.1)
val_datagen = ImageDataGenerator(preprocess_input)

# 训练数据生成器
train_generator = train_datagen.flow_from_directory(
    'path_to_train_data',
    target_size=(224, 224),
    batch_size=32,
    class_mode='binary'
)

# 验证数据生成器
validation_generator = val_datagen.flow_from_directory(
    'path_to_val_data',
    target_size=(224, 224),
    batch_size=32,
    class_mode='binary'
)

# 训练模型
history = model.fit(
    train_generator,
    steps_per_epoch=len(train_generator),
    epochs=10,
    validation_data=validation_generator,
    validation_steps=len(validation_generator)
)