test: Add tests for save_for_aiu functionality w/ tiny models#126
Merged
BrandonGroth merged 16 commits intoJun 9, 2025
Conversation
BrandonGroth
requested review from
andrea-fasoli,
chichun-charlie-liu,
kcirred,
nwang-ibm and
tharapalanivel
as code owners
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
…g, and changed qconfig_save warning to logger.info Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
…tting Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
BrandonGroth
commented
Jun 4, 2025
| else: | ||
| self.register_buffer( | ||
| "clip_val", torch.zeros(perCh) if perCh else torch.Tensor([1.0]) | ||
| "clip_val", torch.zeros(perCh) if perCh else torch.Tensor([0.0]) |
Collaborator
Author
There was a problem hiding this comment.
Uninitialized clip_vals are supposed to be zero tensors, but for some reason this one was set to 1.0. This is consistent for other quantizers currently.
BrandonGroth
commented
Jun 4, 2025
| dataset, | ||
| batch_size=batch_size, | ||
| shuffle=False, | ||
| collate_fn=lambda batch: tuple(torch.stack(samples) for samples in zip(*batch)), |
Collaborator
Author
There was a problem hiding this comment.
This collate_fn was needed to return a tuple(tensor, tensor) as stated in the comment above.
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
Signed-off-by: Brandon Groth <brandon.m.groth@gmail.com>
andrea-fasoli
approved these changes
Jun 6, 2025
andrea-fasoli
left a comment
andrea-fasoli
left a comment
Collaborator
There was a problem hiding this comment.
these unit tests for save_for_aiu look good to me now!
| Returns: | ||
| dict: Model state dictionary | ||
| """ | ||
| return torch.load(fname, weights_only=True) |
Collaborator
There was a problem hiding this comment.
do we need a func for a one-liners?
| """ | ||
| Delete any known files lingering before starting test | ||
| """ | ||
| delete_file("qcfg.json") |
Collaborator
There was a problem hiding this comment.
seems like the purpose is to auto-apply this init-like func before every tests? maybe the we can rename it to something more generic, so we could do more than just delete files if needed?
| """ | ||
| Delete any known files lingering before starting test | ||
| """ | ||
| delete_file("qcfg.json") |
Collaborator
There was a problem hiding this comment.
same as above
|
|
||
|
|
||
| @pytest.fixture(autouse=True) | ||
| def delete_files(): |
6 tasks
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Description of the change
Added a new tests/models named test_save_aiu.py. This includes implementing new "tiny" models for BERT, Llama, and Granite from custom configs that are trimmed down.
Also, several test files have been cleaned up with the new autouse=True fixture for file deletion before a test starts.
How to verify the PR
Change to tests/models and run
pytest test_save_aiu.pyWas the PR tested