微调、迁移学习

查看神经网络层

from torchvision.models import vgg16,vgg16_bn
model = vgg16_bn(pretrained=True)
for k,v in model.named_parameters():
    print(k)

打印输出

features.0.weight
features.0.bias
features.1.weight
features.1.bias
features.3.weight
features.3.bias
features.4.weight
features.4.bias
features.7.weight
features.7.bias
features.8.weight
features.8.bias
features.10.weight
features.10.bias
features.11.weight
features.11.bias
features.14.weight
features.14.bias
features.15.weight
features.15.bias
features.17.weight
features.17.bias
features.18.weight
features.18.bias
features.20.weight
features.20.bias
features.21.weight
features.21.bias
features.24.weight
features.24.bias
features.25.weight
features.25.bias
features.27.weight
features.27.bias
features.28.weight
features.28.bias
features.30.weight
features.30.bias
features.31.weight
features.31.bias
features.34.weight
features.34.bias
features.35.weight
features.35.bias
features.37.weight
features.37.bias
features.38.weight
features.38.bias
features.40.weight
features.40.bias
features.41.weight
features.41.bias
classifier.0.weight
classifier.0.bias
classifier.3.weight
classifier.3.bias
classifier.6.weight
classifier.6.bias

前面的feature为特征提取层，后面的classifier为分类头，训练时，可以冻结前面层的参数不进行训练，（梯度设置为false），将classifier的梯度设置为true进行训练（微调、迁移学习）