CocoZou1994

Nice Work! 为了方便进行对比实验，麻烦发一份训练好的权重模型到本人邮箱，我的邮箱是[email protected] 不胜感激！

Whether your data has been preprocessed, please refer to the code for preprocessing “Follow the python3 data/preprocessBraTS.py which is referenced from the [TransBTS](https://github.com/Wenxuan-1119/TransBTS/blob/main/data/preprocess.py)”

You should carefully check your input data. More details could you provide? 在 2022-09-26 21:46:10，"Xiuxiu21" ***@***.***> 写道： 我已按照您的方法处理数据，但还是有这个问题 — Reply to this email directly, view it on GitHub, or unsubscribe....

There is a high probability that there is a problem with the definition of “temp”, please check it carefully. The data is relatively large, if others have this problem, I...

you should check the input data, detail about the Lines 213-248 "sample = transform(sample)"

Sorry to make you confused. We have check the codes on the journal version. please check in follow link. https://github.com/Cocofeat/UMIS/blob/31fef3f679c9c4dc5def1c4ec1e204a22e446b52/trainUMIS.py#L283 Note: This is because we only used 2 modal inputs...

Sorry, we don't have the valid dataset. ---- Replied Message ---- | From | ***@***.***> | | Date | 08/01/2023 16:54 | | To | Cocofeat/TBraTS ***@***.***> | | Cc...

ice and dice loss both decrease, and KL loss is always 0, which seems impossible. It is suggested that you check the code and focus on the log part in...

Is L_KL the value after multiplying by a balance factor of 0.2 or the previous value, it looks a little too small.

Thanks for your attention! line 268 is ok. line 127 is modified