weihuang-cs / IC4SSL

Inherent Consistency for Semi-supervised Medical Image Segmentation
9 stars 1 forks source link

The weights for autoencoder #1

Open Josoen opened 4 days ago

Josoen commented 4 days ago

Nice work for SSMIS :) Could you provide the weights for autoencoder?

    proj_model = net_factory(net_type="aotuencoder", in_chns=4, class_num=num_classes)
    proj_path = os.path.join(f"../assets/ae_{args.labelnum}.pth")
    logging.info(f"Load autoencoder param from: {proj_path}.")
    proj_model.load_state_dict(torch.load(proj_path))
    proj_model.eval()

the ./assert/ folder only have unet weights

mnmnk43434 commented 4 days ago

Hello, we tried to solve the issue.

This is what we did:

Modified the code to use UNet instead of autoencoder, and updated the weight loading process to use available UNet weights.

You can review changes in this commit: https://github.com/mnmnk43434/weihuang-cs-IC4SSL-1/commit/0a850ce908dda30db2ebeac94efce8f82e275f14.

[!CAUTION] Disclaimer: The concept of solution was created by AI and you should never copy paste this code before you check the correctness of generated code. Solution might not be complete, you should use this code as an inspiration only.


If you no longer want Latta AI to attempt solving issues on your repository, you can block this account.