- 28 Apr, 2023 1 commit
-
-
Neta Zmora authored
* iFix LN ONNX export When exporting LayerNorm make sure that the weights and bias inputs have the same type as the LN input. Also: * Add a regression test. * Add environment variable to override directory of generated test artifacts Signed-off-by:
Neta Zmora <nzmora@nvidia.com> * fix envvar Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com> * fix linting Signed-off-by:
-
- 18 Apr, 2023 1 commit
-
-
Tim Moon authored
Signed-off-by:
Tim Moon <tmoon@nvidia.com> Co-authored-by:
-
- 17 Apr, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* Add tests for cuda graph capture Signed-off-by:
-
- 13 Apr, 2023 1 commit
-
-
Neta Zmora authored
* Fix model load exception when state resides on GPU - Whenever converting a torch.tensor to numpy, we need to first migrate the tensor storage to the host CPU. - Add a warning not to do contant-folding when exporting to ONNX. This is due to a torch.onnx export bug. - Refactor compare_outputs Signed-off-by:
-
- 07 Apr, 2023 2 commits
-
-
ngoyal2707 authored
* made bias configurable Signed-off-by:
Naman Goyal <naman@fb.com> * removed commented lines Signed-off-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com> Signed-off-by:
ngoyal2707 <ngoyal2707@users.noreply.github.com> * Update transformer_engine/pytorch/jit.py Co-authored-by:
-
Kirthi Shankar Sivamani authored
* small cleanup before starting Signed-off-by:
-
- 04 Apr, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* Add FP8 support for Ada Signed-off-by:
-
- 03 Apr, 2023 1 commit
-
-
galagam authored
* Bugfix - compute scale_inv when loading checkpoint Signed-off-by:
Gal Hubara Agam <ghubaraagam@nvidia.com> * Save inverse scale in extra state tensor + minor CR fixes Signed-off-by:
-
- 30 Mar, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* Change FP8 recipe defaults Signed-off-by:
-
- 29 Mar, 2023 1 commit
-
-
tcherckez-nvidia authored
Signed-off-by:
Tal Cherckez <tcherckez@nvidia.com> Signed-off-by:
-
- 23 Mar, 2023 1 commit
-
-
Neta Zmora authored
* Fix GELU ONNX export * Wrap GELU export with cast to/from FP32 to achieve same compute precision as TE. * Increase GELU export test thresholds. * Change export to ONNX opset 17 for smaller representation of LN (single node instead of subgraph). * Remove the need for LN work-around for ORT Signed-off-by:
-
- 18 Mar, 2023 1 commit
-
-
Neta Zmora authored
Add an option to serialize test i/o to file Small refactoring of the inferencing code. Change the default directory where generated ONNX files are stored. Use the temp directory to avoid clogging the file system. Add an option to serialize test input/output tensors to a Polygraphy RunResults object. Signed-off-by:
-
- 17 Mar, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* add layernorm1p fp8 test Signed-off-by:
-
- 16 Mar, 2023 1 commit
-
-
Neta Zmora authored
* Add a temporary workaround to layernorm export Seems like ORT is performing template-matching for LN and incorrectly concludes that it doesn't have a kernel for FP32 LN. The work-around adds the addition of fake_zero which is meant to prevent the template matching while keeping the graph virtually unchanged. This also requires `do_constant_folding=False` in `torch.onnx.export`. Signed-off-by:
-
- 11 Mar, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* deprecate qk layer scaling and fp32 softmax args Signed-off-by:
-
- 07 Mar, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* ignore self attention mask for causal type Signed-off-by:
-
- 24 Feb, 2023 1 commit
-
-
Jeng Bai-Cheng authored
* move TE/PyTorch UT to tests/pytorch 1. move tests/* files to tests/pytorch/ 2. adjust UT paths in qa/L0_unittest/test.sh Signed-off-by:
Ryan Jeng <rjeng@nvidia.com> * update build.yml Signed-off-by:
-
