Void

More details result for onnx,check_model: ``` The model is invalid: No Op registered for RotateTRT with domain_version of 13 ==> Context: Bad node spec for node. Name: OpType: RotateTRT ```...

Sorry to say that I don't have enough time to solve this problem, and really hope that you guys can solve it. @serwansj @matcosta23

refer to [dev/1.0/mmdet3d/models/roi_heads/point_rcnn_roi_head/line213](https://github.com/open-mmlab/mmdetection3d/blob/c0c378f2154238a65446f7e72481a2025df4bb4d/mmdet3d/models/roi_heads/point_rcnn_roi_head.py#L213C5-L213C19), I simply add decrator `@torch.no_grad()` for function `_assign_and_sample()`, and the GPU memory usage remains stable during training.

@galeselee , I tried X11 forwarding, but the window is black, have you got this issue? 

@galeselee I tried to run `example/tutorials/visualization.py,` nothing wrong and I can see rendered pictures. **However**, `hello_genesis` is still **black**.....

same proposal for `"isaacsim/exts/omni.isaac.core/omni/isaac/core/utils/random.py"`

Thanks for your reply! I guess I know the main idea of your great work, and I just wonder **_When_** to update important parameters. I mean, is there any possibility...

I have tried to get the mask(important parameters) after the parameter update, here is the result: - **before**: - Number of params: 2.39M - **Number of FLOPs: 283.33M** - Best...