SWIN transformer pretrained model issue

[abhiagwl4262]

trafficstars

About this repo

• commit - https://github.com/wang-xinyu/tensorrtx/commit/0f944b89750c2cd2e6be7933f84d61eb4e34ec83?
• model - SWIN-Transformer

Your problem

python gen_wts.py upernet_swin_tiny_patch4_window7_512x512.pth 
Traceback (most recent call last):
  File "gen_wts.py", line 8, in <module>
    model = torch.load(pt_file, map_location=torch.device('cpu'))['model'].float()  # load to FP32
KeyError: 'model' 

Can you please provide the link to the pre-trained model you used for swin usecase or modify the script such that we are able to use the pretrained model from https://github.com/SwinTransformer/Swin-Transformer-Semantic-Segmentation repo ?

[wang-xinyu]

@wdhao was using a custom model, but the code can still be referenced.

@abhiagwl4262 If you can adapt the code to make it work with the pretrained model, then welcome your PR.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.