-
Notifications
You must be signed in to change notification settings - Fork 10
Convert EKFAC tests to pytest #68
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
Open
smarter
wants to merge
10
commits into
EleutherAI:ekfac
Choose a base branch
from
smarter:ekfac-pytests
base: ekfac
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
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
The keyword parameter `dtype` for AutoModelForCausalLM.from_pretrained does not exist in version 4.54.1 which was present in uv.lock (this parameter used to be called `torch_dtype` which is now a deprecated alias).
Also rework default handling to avoid specifying default values in multiple places.
…ection) This way when we start using pytest, test failures will be properly reported. test_eigenvalue_correction had no explicit criterion for success so I made one up.
df1ef3d to
6928233
Compare
LouisYRYJ
reviewed
Nov 14, 2025
This includes using fixtures for ground truth generation and test configuration, so that we can just do: uv run pytest -sv tests/ekfac_tests and ground truth will be auto-generated.
Ran "uv pre-commit run --all-files" which reads from .pre-commit-config.yaml Unfortunately pre-commit does not respect tool settings in pyproject.toml, so right now there's conflicting informations in pyproject.toml and .pre-commit-config.yaml and so different settings and tool versions used depending on how we run tools.
test_eigenvalue_corrections had to be disabled due to precision errors: h.6.attn.attention.out_proj: max_rel_diff=2.285% h.6.mlp.c_proj: max_rel_diff=3.599% h.7.attn.attention.out_proj: max_rel_diff=4.041% h.7.mlp.c_proj: max_rel_diff=2.204%
It seems the working-directory parameter in the CI config is ignored if pyproject.toml configures pyright, so tweak that instead.
Overwriting is allowed using the --overwrite flag.
c9a5aff to
c61a2e1
Compare
for more information, see https://pre-commit.ci
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.
This includes using fixtures for ground truth generation and test configuration,
so that we can just do:
uv run pytest -sv tests/ekfac_tests
and ground truth will be auto-generated.
Additionally, this required adding a pass/fail threshold to tests/ekfac_tests/test_eigenvalue_correction.py.
I haven't tested test_apply_ekfac yet.