Weihang Xia

ERROR with no trace back.

2

mmdeploy was built on my platform with Windows11, cuda 11.7, cudnn 8.4 tensorrt 8.4.0.6, torch 1.12 This bug occurs while I was converting mask rcnn swin and yolov3 `2022-07-04 19:56:36,687...

Training mod: using ProgressMeter.jl in `train!`

3

Add a progress bar with display of training loss in train!(...). For training less than 1 batch there is no display of progress bar and training loss.

🐛 [Bug] torch.jit.frontend.NotSupportedError: Compiled functions can't take variable number of arguments or use keyword-only arguments with defaults

7

Error: torch.jit.frontend.NotSupportedError: Compiled functions can't take variable number of arguments or use keyword-only arguments with defaults: File "/home/user/anaconda3/envs/torchtensorrt/lib/python3.8/site-packages/torch/utils/checkpoint.py", line 145 def checkpoint(function, *args, **kwargs): ~~~~~~~

Swin Transformer related utility function

6

add a function which returns a matrix of relative position index from Swin Transformer

[BUG]: Errors when compiling numpy-related code on Apple silicon platform

1

### Required prerequisites - [X] Make sure you've read the [documentation](https://pybind11.readthedocs.io). Your issue may be addressed there. - [X] Search the [issue tracker](https://github.com/pybind/pybind11/issues) and [Discussions](https:/pybind/pybind11/discussions) to verify that this hasn't...