Pengfei Guo
Pengfei Guo
**Describe the bug** This Class [ConvertToMultiChannelBasedOnBratsClassesd](https://github.com/Project-MONAI/tutorials/blob/c75cce3887c01fcea0b69e89f4d1e977fae2e782/acceleration/distributed_training/brats_training_ddp.py#L95) is used in [brats_training_ddp.py](https://github.com/Project-MONAI/tutorials/blob/c75cce3887c01fcea0b69e89f4d1e977fae2e782/acceleration/distributed_training/brats_training_ddp.py#L95) and [brats_segmentation_3d.ipynb](https://github.com/Project-MONAI/tutorials/blob/c75cce3887c01fcea0b69e89f4d1e977fae2e782/3d_segmentation/brats_segmentation_3d.ipynb). It may map the label data to wrong classes. According to its docstring, it describes that label 2...
In your paper. you mentioned the networks co-trained in 3DSeg-8 dataset, including Brats. Would like lease more detail about co-training? In other words, Brats dataset contains T1, T2, T1c, Flair...
At /src/models/ gain.py 86 gradient = logits * labels_ohe 87 grad_logits = (logits * labels_ohe).sum() 88 grad_logits.backward(gradient=gradient, retain_graph=True) Line 88 shows an error: File "C:\ProgramData\Anaconda2\envs\python3\lib\site-packages\torch\autograd\__init__.py", line 90, in backward allow_unreachable=True)...
Fixes # . ### Description Add training script for MAISI ControlNet. ### Checks - [ ] Avoid including large-size files in the PR. - [ ] Clean up long text...