Conversation
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
There was a problem hiding this comment.
lintrunner found more than 10 potential problems in the proposed changes. Check the Files changed tab for more details.
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
|
Hi, Sorry to barge in. I was looking into #18313 and was confused as the CPU implementation seemed complete, and this is how I ended up here. IIUC only the CPU implementation is complete? Could I try to contribute towards the CUDA implementation? Just looking to contribute and learn. |
|
@liqunfu Hi ,Do you have any plan to implement a CUDA version ? |
|
After torch.nn.functional.grid_sample is exported to onnx, it is very slow compared to torch. Are there plans to improve the performance of grid_sample in onnxruntime-gpu? |
Agreed. |
…operator (#27201) ### Description 1. Supports volumetric input grid sampling in the CUDA EP `GridSample` operator (i.e.) 5-D input tensor a.k.a 3-D spatial data 2. Registers the CUDA `GridSample` operator for opsets 20 and 22 3. Supports both NCHW and NHWC layouts for volumetric inputs 4. Does not support `cubic` mode for volumetric inputs for now and this is consistent with the CPU version of the implementation and hence will not cause "functional regression" (i.e.) `cubic` mode for 3-D spatial data is not supported on CPU and CUDA before and after this change. This is a TODO for the future. 5. There are enough unit tests in `grid_sample_test.cc` to cover the volumetric input case and this is run in both NCHW (NCDHW for volumetric case) and NHWC (NDHWC for volumetric case) layouts for the CUDA EP ### Motivation and Context Resolve #21382 Resolve #18942 Resolve #16581 Resolve #18313 Related CPU PRs (for opset 20 and opset 22): #17744 && #23344
8 participants
Description
gridsample has been update in ONNX 20. update cpu provider to it
Motivation and Context
prepare for ORT 1.17.0