Ke Wang

One way is to look at the losses, the other way is to do the patch retrieval experiments described in the paper.

Please make sure your BART works fine. Also, I don't think our code supports single-coil, it can be easily implemented by getting rid of all sensmaps related stuff. Hope it...

High, thanks for your question! The training time is inherently long due to the large datasets and unrolled networks. In terms of GPU memory, in our paper, we used the...

Please set args.loss_normalized = True and try again, and I will solve this issue, Thanks!

I see, how did you train the UFLoss? the error is not about the normalization. It's about the checkpoint loading, how many patches did you use to train the UFLoss...

Hi, I believe it is a sigpy version mismatch! Maybe we can schedule a quick chat to address these issues? And I will update the repo accordingly. Apologize for the...