Models.features , model.encoder, and model.decoder

vision
1

I see a lot of examples using model.features

torchvision.models.vgg16(pretrained=pretrained).features

so, can we use model.encoder or model.decoder too? or it depends on the architecture of the model? where can we found the documentation of them?

2

It depends on the architecture of the model and you can check the internal modules in the model definition.
E.g. torchvision.models are implemented here.