Streamline KD & QAD transformers Trainers #708
Merged
+88
−177
Conversation
This file contains hidden or 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
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #708 +/- ##
==========================================
- Coverage 74.69% 74.62% -0.07%
==========================================
Files 192 192
Lines 18946 18989 +43
==========================================
+ Hits 14152 14171 +19
- Misses 4794 4818 +24 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
AAnoosheh
requested review from
ChenhanYu,
kevalmorabia97,
kinjalpatel27,
mxinO and
realAsma
examples/llm_qat/README.md
Outdated
| > **_NOTE:_** `launch.sh` defaults to use `LlamaDecoderLayer` as the transformer layer class. If your model uses a different class, you need to pass `--fsdp_transformer_layer_cls_to_wrap <your_layer_class>` to the `launch.sh` script. For example, for `Qwen/Qwen3-8B`, specify `--fsdp_transformer_layer_cls_to_wrap Qwen3DecoderLayer` as an additional argument. | ||
| > **_NOTE:_** The script defaults to using FSDP1. To use FSDP2, pass "--use_fsdp2 True" to the `launch.sh` script. Note that FSDP2 is less stable than FSDP1 currently. Use it with caution. | ||
| > **_NOTE:_** The script defaults to using FSDP1. To use FSDP2, pass "--backend=fsdp2" to the `launch.sh` script. Note that FSDP2 is less stable than FSDP1 currently. Use it with caution. |
Collaborator
There was a problem hiding this comment.
Is this statement still valid? Note that FSDP2 is less stable than FSDP1 currently. Use it with caution.
Contributor
Author
There was a problem hiding this comment.
I doubt it, but I don't have proof. I don't have proof that it is less stable either.
kevalmorabia97
approved these changes
Dec 19, 2025
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
AAnoosheh
force-pushed
the
aanoosheh/kd-trainer-streamline
branch
from
bde5788 to
190e4d2
Compare
realAsma
reviewed
Dec 24, 2025
realAsma
reviewed
Dec 24, 2025
AAnoosheh
force-pushed
the
aanoosheh/kd-trainer-streamline
branch
from
c4c0d19 to
9f3b0f8
Compare
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
AAnoosheh
force-pushed
the
aanoosheh/kd-trainer-streamline
branch
from
9f3b0f8 to
50cb0f4
Compare
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
AAnoosheh
force-pushed
the
aanoosheh/kd-trainer-streamline
branch
from
f6e1196 to
796b023
Compare
realAsma
reviewed
Jan 7, 2026
realAsma
reviewed
Jan 7, 2026
realAsma
reviewed
Jan 7, 2026
realAsma
reviewed
Jan 7, 2026
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
realAsma
reviewed
Jan 8, 2026
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
realAsma
reviewed
Jan 9, 2026
realAsma
reviewed
Jan 9, 2026
realAsma
reviewed
Jan 9, 2026
realAsma
reviewed
Jan 9, 2026
| # Note: QAD doesn't work with FSDP wrapped model. We quantize model before the wrapper. | ||
| # The drawback is that we can't train a model that is bigger than a single GPU memory. | ||
| # And memory efficient loading doesn't work. | ||
| # Note: FSDP memory efficient loading doesn't work. |
Contributor
There was a problem hiding this comment.
wont be needed if we do https://github.com/NVIDIA/Model-Optimizer/pull/708/files#r2677776721
Suggested change
| # Note: FSDP memory efficient loading doesn't work. |
realAsma
reviewed
Jan 9, 2026
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
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?
Type of change: ? Refactor and stabilization
Overview:
Usage
# Add a code snippet demonstrating how to use thisTesting
Before your PR is "Ready for review"
Additional Information