-
Notifications
You must be signed in to change notification settings - Fork 2.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Load lang id any2any training mengruw #8961
Load lang id any2any training mengruw #8961
Commits on May 17, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 4084669 - Browse repository at this point
Copy the full SHA 4084669View commit details
Commits on May 19, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 1dc8b37 - Browse repository at this point
Copy the full SHA 1dc8b37View commit details
Commits on May 23, 2023
-
Fix k2 installation in Docker with CUDA 12 (NVIDIA#6707)
Signed-off-by: Vladimir Bataev <vbataev@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 0ca1dd3 - Browse repository at this point
Copy the full SHA 0ca1dd3View commit details
Commits on May 24, 2023
-
Signed-off-by: smajumdar <titu1994@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for db6e29b - Browse repository at this point
Copy the full SHA db6e29bView commit details
Commits on May 26, 2023
-
VP Fixes for converter + Config management (NVIDIA#6698) (NVIDIA#6738)
* [Temp] VP Fixes Signed-off-by: smajumdar <titu1994@gmail.com> * Revert logging Signed-off-by: smajumdar <titu1994@gmail.com> --------- Signed-off-by: smajumdar <titu1994@gmail.com> (cherry picked from commit b6f46a0)
Configuration menu - View commit details
-
Copy full SHA for 2e2df4a - Browse repository at this point
Copy the full SHA 2e2df4aView commit details -
Fix fastpitch test nightly (NVIDIA#6742)
Signed-off-by: hsiehjackson <c2hsieh@ucsd.edu>
Configuration menu - View commit details
-
Copy full SHA for 4df8f33 - Browse repository at this point
Copy the full SHA 4df8f33View commit details -
check for first or last stage (NVIDIA#6708)
* check for first or last stage Signed-off-by: ericharper <complex451@gmail.com> * remove redundant check Signed-off-by: ericharper <complex451@gmail.com> * fix typo Signed-off-by: ericharper <complex451@gmail.com> * add map_location Signed-off-by: ericharper <complex451@gmail.com> --------- Signed-off-by: ericharper <complex451@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for e806e11 - Browse repository at this point
Copy the full SHA e806e11View commit details
Commits on May 29, 2023
-
Bug fix to restore act ckpt (NVIDIA#6753)
* Bug fix to restore act ckpt Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --------- Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for dbd6a56 - Browse repository at this point
Copy the full SHA dbd6a56View commit details
Commits on May 31, 2023
-
Bug fix to reset sequence parallelism (NVIDIA#6756)
* Bug fix to reset sequence parallelism Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com> * Update seq par reset/restore Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com> * Add nested loop Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com> --------- Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for a0f757e - Browse repository at this point
Copy the full SHA a0f757eView commit details -
Fix checkpointed forward and add test for full activation checkpointi…
…ng (NVIDIA#6744) * fix checkpointed forward and add test for full activation checkpointing Signed-off-by: Abhinav Khattar <aklife97@gmail.com> * add method Signed-off-by: Abhinav Khattar <aklife97@gmail.com> * add method Signed-off-by: Abhinav Khattar <aklife97@gmail.com> --------- Signed-off-by: Abhinav Khattar <aklife97@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 39dd654 - Browse repository at this point
Copy the full SHA 39dd654View commit details -
Signed-off-by: smajumdar <titu1994@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 216bcab - Browse repository at this point
Copy the full SHA 216bcabView commit details
Commits on Jun 1, 2023
-
add call to p2p overlap (NVIDIA#6779)
* add call to p2p overlap Signed-off-by: Abhinav Khattar <aklife97@gmail.com> * update Jenkins for test Signed-off-by: Abhinav Khattar <aklife97@gmail.com> --------- Signed-off-by: Abhinav Khattar <aklife97@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 4ecc769 - Browse repository at this point
Copy the full SHA 4ecc769View commit details -
Fix get_parameters when using main params optimizer (NVIDIA#6764)
* fix get param Signed-off-by: ericharper <complex451@gmail.com> * change name Signed-off-by: ericharper <complex451@gmail.com> --------- Signed-off-by: ericharper <complex451@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 1486b12 - Browse repository at this point
Copy the full SHA 1486b12View commit details -
* initial POC for LDDL Bert * Finish LDDL POC * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * address comments * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix merge head * resolving merge * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * add support for val/test loaders * change to new LDDL class + add winding * fix logging level * fix winding * test fix * fixes to winding * add file system * add prepemption optimizations * more logging * more prints * better logging * asfsf * add barrier * removing prints * working with mb lddl loader * final changes * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * update requirements file with LDDL Signed-off-by: wdykas <wdykas@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * revert adding to requirements --------- Signed-off-by: wdykas <wdykas@nvidia.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Eric Harper <complex451@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for aff5217 - Browse repository at this point
Copy the full SHA aff5217View commit details -
Debug Transformer Engine FP8 support with Megatron-core infrastructure (
NVIDIA#6740) * Construct FP8 amax reduction group Signed-off-by: Tim Moon <tmoon@nvidia.com> * update core for CI Signed-off-by: Abhinav Khattar <aklife97@gmail.com> --------- Signed-off-by: Tim Moon <tmoon@nvidia.com> Signed-off-by: Abhinav Khattar <aklife97@gmail.com> Co-authored-by: Abhinav Khattar <aklife97@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 4bbb3c6 - Browse repository at this point
Copy the full SHA 4bbb3c6View commit details -
Tensor-parallel communication overlap with userbuffer backend (NVIDIA…
…#6780) * add interfaces for tp_communication overlap [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Interface to provide custom userbuffer communicator settings by yaml file [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Construct MPI process group for userbuffers support Signed-off-by: Tim Moon <tmoon@nvidia.com> --------- Signed-off-by: Tim Moon <tmoon@nvidia.com> Co-authored-by: Tim Moon <tmoon@nvidia.com> Co-authored-by: Abhinav Khattar <aklife97@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for e4460d1 - Browse repository at this point
Copy the full SHA e4460d1View commit details
Commits on Jun 2, 2023
-
Fix adapter tutorial r1.19.0 (NVIDIA#6776)
* Fix TTS adapter tutorial Signed-off-by: hsiehjackson <c2hsieh@ucsd.edu> * Fix version Signed-off-by: hsiehjackson <c2hsieh@ucsd.edu> --------- Signed-off-by: hsiehjackson <c2hsieh@ucsd.edu>
Configuration menu - View commit details
-
Copy full SHA for 9bd8ecd - Browse repository at this point
Copy the full SHA 9bd8ecdView commit details -
Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>
Configuration menu - View commit details
-
Copy full SHA for 913e5e5 - Browse repository at this point
Copy the full SHA 913e5e5View commit details -
Bug fix for reset_sequence_parallel_args (NVIDIA#6802)
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for a8aa8f1 - Browse repository at this point
Copy the full SHA a8aa8f1View commit details
Commits on Jun 5, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 0e0253e - Browse repository at this point
Copy the full SHA 0e0253eView commit details
Commits on Jun 6, 2023
-
update core version (NVIDIA#6817)
Signed-off-by: Abhinav Khattar <aklife97@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 41bb941 - Browse repository at this point
Copy the full SHA 41bb941View commit details -
Add trainer.validate example for GPT (NVIDIA#6794)
* add trainer.validate example Signed-off-by: ericharper <complex451@gmail.com> * clean up white space Signed-off-by: ericharper <complex451@gmail.com> * add mbs and gbs to the config Signed-off-by: ericharper <complex451@gmail.com> --------- Signed-off-by: ericharper <complex451@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 45144f5 - Browse repository at this point
Copy the full SHA 45144f5View commit details
Commits on Jun 8, 2023
-
fix notebook error (NVIDIA#6840)
Signed-off-by: Yi Dong <yidong@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for dc52b94 - Browse repository at this point
Copy the full SHA dc52b94View commit details -
Configuration menu - View commit details
-
Copy full SHA for 4239b80 - Browse repository at this point
Copy the full SHA 4239b80View commit details
Commits on Jun 13, 2023
-
Add API docs for NeMo Megatron (NVIDIA#6850)
* add model pretraining and customization classes Signed-off-by: ericharper <complex451@gmail.com> * fix Signed-off-by: ericharper <complex451@gmail.com> * test width Signed-off-by: ericharper <complex451@gmail.com> * increase middle pane width Signed-off-by: ericharper <complex451@gmail.com> * add modules and datasets Signed-off-by: ericharper <complex451@gmail.com> * remove global in t5 dataset s and fix formatting in megatron base model Signed-off-by: ericharper <complex451@gmail.com> --------- Signed-off-by: ericharper <complex451@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 87e1b81 - Browse repository at this point
Copy the full SHA 87e1b81View commit details
Commits on Jun 14, 2023
-
Apply garbage collection interval to validation steps (NVIDIA#6870)
* Apply garbage collection inverval to validation steps Signed-off-by: Sangkug Lym <slym@nvidia.com> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --------- Signed-off-by: Sangkug Lym <slym@nvidia.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for f875702 - Browse repository at this point
Copy the full SHA f875702View commit details
Commits on Jun 15, 2023
-
update mcore version (NVIDIA#6875)
Signed-off-by: ericharper <complex451@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 2331b06 - Browse repository at this point
Copy the full SHA 2331b06View commit details
Commits on Apr 17, 2024
-
fix language id issue for any2any training. This fix loads language i…
…ds from .yaml file
Dockerfile builder committedApr 17, 2024 Configuration menu - View commit details
-
Copy full SHA for f0fa541 - Browse repository at this point
Copy the full SHA f0fa541View commit details