Add ability to enable/disable act ckpt and seq parallelism in GPT #6327
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
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
for more information, see https://pre-commit.ci
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
|
This looks good to me and I've been using this for SFT models to turn this on/off between training and validation. I'll let @ericharper do a final review. |
ericharper
reviewed
Apr 6, 2023
nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py
Outdated
Show resolved
Hide resolved
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
for more information, see https://pre-commit.ci
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
for more information, see https://pre-commit.ci
ericharper
reviewed
Apr 11, 2023
nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py
Outdated
Show resolved
Hide resolved
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
for more information, see https://pre-commit.ci
ericharper
reviewed
Apr 11, 2023
nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py
Outdated
Show resolved
Hide resolved
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
…t function. Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com>
for more information, see https://pre-commit.ci
ericharper
approved these changes
Apr 13, 2023
hsiehjackson
pushed a commit
to hsiehjackson/NeMo
that referenced
this pull request
Jun 2, 2023
…IDIA#6327) * Add ability to enable/disable act ckpt and seq parallelism 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 * Remove num_micro_batches_with_partial_activation_checkpoints Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com> * Added property to self.model and added restore/reset config values. 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 * Use self.model property 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 * Removed original_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 * Add docstrings to reset/restore act ckpt Signed-off-by: Markel Sanz Ausin <markelsanz14@gmail.com> * Property removed from self.model and replaced with get_gpt_module_list function. 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> Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com> Signed-off-by: hsiehjackson <c2hsieh@ucsd.edu>
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.
What does this PR do ?
Adds functions to enable and disable activation checkpointing and sequence parallelism. This way, we can enable them during training and disable them during inference/generation.
Collection: nemo/collections/nlp/language_modeling
Changelog
Usage
# Add a code snippet demonstrating how to use thisBefore your PR is "Ready for review"
Pre checks:
PR Type:
If you haven't finished some of the above items you can still open "Draft" PR.
Who can review?
Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.
Additional Information