fix npu bug. #3136
Open
fix npu bug. #3136
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
HAOCHENYE
reviewed
Jul 16, 2024
| if torch.cuda.current_device() != points_device: | ||
| torch.cuda.set_device(points_device) | ||
| elif points.device.type == 'npu': | ||
| boxes[:, :, 2] += boxes[:, :, 5] / 2.0 |
There was a problem hiding this comment.
Could we move this into npu kernel?
Comment on lines
+40
to
+58
| if ctx.device == 'npu': | ||
| import ads_c | ||
| voxel_idx = ads_c.point_to_voxel(coors, None, None) | ||
| unique_res = ads_c.unique_voxel(voxel_idx) | ||
| num_voxels, uniqued_voxel_idx, prefix_sum, \ | ||
| argsort_coor = unique_res | ||
| voxel_coors = ads_c.voxel_to_point(uniqued_voxel_idx, None, None) | ||
| voxel_feats, \ | ||
| compare_mask = ads_c.npu_dynamic_scatter(feats, coors, | ||
| prefix_sum, | ||
| argsort_coor, | ||
| num_voxels, | ||
| reduce_type) | ||
| ctx.reduce_type = reduce_type | ||
| ctx.feats_shape = feats.shape | ||
| ctx.save_for_backward(prefix_sum, argsort_coor, compare_mask) | ||
| ctx.mark_non_differentiable(voxel_coors) | ||
| return voxel_feats, voxel_coors | ||
|
|
| @@ -50,6 +69,19 @@ def forward(ctx: Any, | |||
| def backward(ctx: Any, | |||
| grad_voxel_feats: torch.Tensor, | |||
| grad_voxel_coors: Optional[torch.Tensor] = None) -> tuple: | |||
| if ctx.device == 'npu': | |||
| @@ -34,6 +34,21 @@ def forward(ctx: Any, target: torch.Tensor, | |||
|
|
|||
| B, N, _ = target.size() | |||
| m = source.size(1) | |||
| if source.device.type == 'npu': | |||
| # strict to fp32 | |||
| source = source.transpose(2, 1).contiguous() | |||
Comment on lines
+418
to
+419
| dets_cw = dets_cw.float() | ||
| scores = scores.float() |
There was a problem hiding this comment.
Could we move this into npu kernel?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.
Motivation
Please describe the motivation of this PR and the goal you want to achieve through this PR.
Modification
Please briefly describe what modification is made in this PR.
BC-breaking (Optional)
Does the modification introduce changes that break the backward-compatibility of the downstream repositories?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.
Use cases (Optional)
If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.
Checklist
Before PR:
After PR: