how to use pretrained models?

[AI-liu]

[cardwing]

In extractors.py, you can find the pre-trained weights in the following URLs: model_urls = { 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', } Hope this issue can be closed.

[Primus-zhao]

@cardwing when i use resnet pretrained model, it always give tips that lots of layers' weights not given

[alialamiidrissi]

I have the same problem, I get multiple errors when trying to load the pre-trained resnets. It seems that the checkpoint and the created model layers names and shapes don't match.

EDIT: I solved the issue by changing the function 'load_weights_sequential' as follow:

def load_weights_sequential(target, source_state):
    model_to_load= {k: v for k, v in source_state.items() if k in target.state_dict().keys()}
    target.load_state_dict(model_to_load)

[engharat]

are those pretrained weights obtained by training on ImageNet?

[tsing90]

@alialamiidrissi thanks for your solution, worked!

[fasogbon]

Hello, does anybody have a trained model and/or test script to test the method? Thanks

[gaohuiluo]

@alialamiidrissi thank you for your solution. It woked!