[QEff. Finetuning]: Enhance test cases to match intermediate step level loss/metrics #531
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
quic-akuruvil
force-pushed
the
test_cases
branch
from
c9ada4a to
084bb38
Compare
quic-swatia
reviewed
Aug 5, 2025
quic-akuruvil
requested review from
quic-rishinr,
ochougul,
quic-hemagnih and
quic-amitraj
as code owners
quic-akuruvil
force-pushed
the
test_cases
branch
2 times, most recently
from
d76ef7f to
c31b88a
Compare
quic-meetkuma
requested changes
Aug 14, 2025
quic-meetkuma
approved these changes
Aug 14, 2025
quic-swatia
reviewed
Aug 14, 2025
tests/finetune/test_finetune.py
Outdated
| f"{name} length mismatch for scenario '{scenario_key}' (WS: {current_world_size}, Rank: {current_rank}). " | ||
| f"Expected {len(ref_list)} elements, but got {len(actual_list)}." | ||
| ) | ||
| max_diff = np.max(np.abs(np.array(ref_list) - np.array(actual_list))) |
There was a problem hiding this comment.
In case of mismatch, it will report the max diff. It should instead report: 1) The step numbers at which deviation is happening, 2) diff in value at each of these steps. np.isclose() will help in getting the deviated indices. Before this, np.allclose() can be used to check if the assertion is passing or not.
There was a problem hiding this comment.
Added step wise details for deviation
quic-swatia
reviewed
Aug 14, 2025
| REFERENCE_DATA = { | ||
| # Scenario 1: Single-device llama 3.2-1B training on Alpaca dataset. | ||
| "llama_config_alpaca_single_device": { | ||
| "description": "Baseline for Llama on Alpaca single-device", |
There was a problem hiding this comment.
Please add the complete model ID here and in other configs as well.
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
Signed-off-by: Ann Kuruvilla <[email protected]>
quic-akuruvil
force-pushed
the
test_cases
branch
from
4b6e21d to
77106f6
Compare
quic-swatia
approved these changes
Aug 18, 2025
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.
Enable test cases for Intermediate step level loss/metric matching in single and DDP set up.
Nested dictionary structure for mapping the reference losses at different test scenarios. The test scenarios with the ref values are listed in a separate reference file.
The test scenarios at present include single device testing for below models:
Llama, Bert on Alpaca and GSM8k dataset.
REFERNCE DATA based on SDK - 1.21.0.23