- 17 Apr, 2023 3 commits
-
-
Przemek Tredak authored
Signed-off-by:Przemek Tredak <ptredak@nvidia.com>
-
Kirthi Shankar Sivamani authored
* Add tests for cuda graph capture Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com> * add sanity test and address reviews Signed-off-by:
-
Kirthi Shankar Sivamani authored
* use upstream flash-attn Signed-off-by:
-
- 15 Apr, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* Fix gencode flags based on cuda version Signed-off-by:
-
- 14 Apr, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
Signed-off-by:
-
- 13 Apr, 2023 3 commits
-
-
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:
Neta Zmora <nzmora@nvidia.com> * Onnx export: Improve remark text Signed-off-by:
-
zlsh80826 authored
* Add zero_center_gamma/functional pass Signed-off-by:
Reese Wang <rewang@nvidia.com> * Add zero_centered_gamma for fp8_ln_mlp Signed-off-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com>
-
Kaixi Hou authored
Remove the autocast_variable Signed-off-by:kaixih <kaixih@nvidia.com>
-
- 11 Apr, 2023 1 commit
-
-
Trevor Morris authored
* Fix pybind11 install doc Signed-off-by:
Trevor Morris <tmorris@nvidia.com> * Set CMAKE_PREFIX_PATH for TF to find pybind11 Signed-off-by:
-
- 08 Apr, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
Fix cyclic import error in TF Signed-off-by:
-
- 07 Apr, 2023 5 commits
-
-
Kirthi Shankar Sivamani authored
fix nightly docs Signed-off-by: -
ngoyal2707 authored
* made bias configurable Signed-off-by:
Naman Goyal <naman@fb.com> * removed commented lines Signed-off-by:
ngoyal2707 <ngoyal2707@users.noreply.github.com> * Update transformer_engine/pytorch/jit.py Co-authored-by:
-
Ming-Xu Huang authored
* Rename enable_fp8 to is_fp8_enabled. Signed-off-by:
Ming Huang <mingh@nvidia.com> * Adding an API to get an instance of DelayedScaling which is set via fp8_autocast. Signed-off-by:
-
Tim Moon authored
* GitHub actions for linting JAX code Pylint is disabled since JAX container is not publicly available. Signed-off-by:
Tim Moon <tmoon@nvidia.com> * GitHub action for building JAX code Disabled since JAX container is not publicly available. Signed-off-by:
-
Kirthi Shankar Sivamani authored
* small cleanup before starting Signed-off-by:
-
- 05 Apr, 2023 3 commits
-
-
Frédéric Bastien authored
* Add the link to the examples and the development user guide. Signed-off-by:
Frederic Bastien <fbastien@nvidia.com> * Update README.rst Co-authored-by:
Frédéric Bastien <frederic.bastien@gmail.com> --------- Signed-off-by:
-
Frédéric Bastien authored
* Update installation instructio for JAX and add some depenencies. Signed-off-by:
-
Sangkug Lym authored
* async amax reduction add env knob to enable async amax reduction Signed-off-by:
slym <slym@login-preos01.a51.clusters.nvidia.com> * Style fixes 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 2 commits
-
-
Kirthi Shankar Sivamani authored
* Fix segfault during GeLU export Signed-off-by:
-
Kirthi Shankar Sivamani authored
* Change FP8 recipe defaults Signed-off-by:
-
- 29 Mar, 2023 2 commits
-
-
tcherckez-nvidia authored
Signed-off-by:
Tal Cherckez <tcherckez@nvidia.com> Signed-off-by:
-
Ming-Xu Huang authored
* Support transpose_bs when decoded=True Signed-off-by:
-
- 28 Mar, 2023 5 commits
-
-
vasunvidia authored
* Add support for fp8 GEMM BIAS AUX GELU fusion Signed-off-by:
Vasudevan Rengasamy <vrengasamy@nvidia.com> * Fix Lint error Signed-off-by:
-
Jeng Bai-Cheng authored
* refactor JAX examples Signed-off-by:
Ryan Jeng <rjeng@nvidia.com> * fix doc-string Signed-off-by:
-
Trevor Morris authored
* Add tensorflow build Improve build instructions Fix pybind enum usage Fix Python_EXECUTABLE cmake var Move scale_inv calculations to FW Signed-off-by:
-
Tim Moon authored
* Remove zombie process from querying TE install path Co-authored-by:
-
Kirthi Shankar Sivamani authored
* fix usage of return_bias argument 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:
-
- 22 Mar, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
FA doesn't support compute 8.6 with head_dim>64 Signed-off-by:
-
- 21 Mar, 2023 2 commits
-
-
Przemyslaw Tredak authored
Signed-off-by: -
vasunvidia authored
* Initial commit for fp8_transpose_dbias kernel 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 4 commits
-
-
Przemek Tredak authored
Signed-off-by: -
Tim Moon authored
Signed-off-by: -
Przemek Tredak authored
Signed-off-by: -
Kirthi Shankar Sivamani authored
* add layernorm1p fp8 test Signed-off-by:
-
- 16 Mar, 2023 2 commits
-
-
Kirthi Shankar Sivamani authored
Signed-off-by: -
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:
-
