基于tensorflow迁移学习resnet50完整代码

当使用TensorFlow进行迁移学习时，可以使用预训练的ResNet-50模型作为基础模型，并在其基础上进行微调。以下是一个基于TensorFlow的完整代码示例：

import tensorflow as tf
from tensorflow import keras

# 加载预训练模型
base_model = keras.applications.ResNet50(weights='imagenet', include_top=False, input_shape=(224, 224, 3))

# 冻结预训练模型的权重
for layer in base_model.layers:
    layer.trainable = False

# 添加自定义的分类器层
x = base_model.output
x = keras.layers.GlobalAveragePooling2D()(x)
x = keras.layers.Dense(256, activation='relu')(x)
predictions = keras.layers.Dense(num_classes, activation='softmax')(x)

# 创建完整的模型
model = keras.models.Model(inputs=base_model.input, outputs=predictions)

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

# 数据准备和增强
train_datagen = keras.preprocessing.image.ImageDataGenerator(
    rescale=1./255,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True)

test_datagen = keras.preprocessing.image.ImageDataGenerator(rescale=1./255)

train_generator = train_datagen.flow_from_directory(
    train_data_dir,
    target_size=(224, 224),
    batch_size=batch_size,
    class_mode='categorical')

validation_generator = test_datagen.flow_from_directory(
    validation_data_dir,
    target_size=(224, 224),
    batch_size=batch_size,
    class_mode='categorical')

# 训练模型
model.fit(
    train_generator,
    steps_per_epoch=train_steps_per_epoch,
    epochs=num_epochs,
    validation_data=validation_generator,
    validation_steps=validation_steps_per_epoch)

# 评估模型
test_generator = test_datagen.flow_from_directory(
    test_data_dir,
    target_size=(224, 224),
    batch_size=batch_size,
    class_mode='categorical',
    shuffle=False)

scores = model.evaluate(test_generator)
print(f'Test loss: {scores[0]}')
print(f'Test accuracy: {scores[1]}')

请注意，上述代码中的数据集路径、类别数量、训练轮数、批次大小等需要根据您的具体情况进行适当修改。此代码将在训练集上微调预训练的ResNet-50模型，并在验证集和测试集上评估性能。