Add CPU quality regression tests (ResNet50, Qwen3-0.6B, Whisper-tiny.en)#177
Open
roberto-laudani wants to merge 4 commits into
Open
Add CPU quality regression tests (ResNet50, Qwen3-0.6B, Whisper-tiny.en)#177roberto-laudani wants to merge 4 commits into
roberto-laudani wants to merge 4 commits into
Conversation
roberto-laudani
requested review from
Groverkss,
MaheshRavishankar and
kuhar
as code owners
Signed-off-by: Roberto Laudani <laudani@roofline.ai>
roberto-laudani
force-pushed
the
add-resnet50-regression
branch
from
4f4a653 to
928276b
Compare
Contributor
Author
|
The red "Test Torch Models" check is the pre-existing |
roberto-laudani
changed the title
Signed-off-by: Roberto Laudani <laudani@roofline.ai>
roberto-laudani
changed the title
roberto-laudani
force-pushed
the
add-resnet50-regression
branch
from
d9e707d to
ffe38ee
Compare
Signed-off-by: Roberto Laudani <laudani@roofline.ai>
roberto-laudani
force-pushed
the
add-resnet50-regression
branch
from
ffe38ee to
9fb9b06
Compare
chrsmcgrr
reviewed
Jun 25, 2026
chrsmcgrr
left a comment
chrsmcgrr
left a comment
There was a problem hiding this comment.
Looks good! Would it also be possible to have the binary files on HF? We would want to avoid checking in binary files because they can't be diffed. Changing it later will cause a new copy in the history.
| args = parser.parse_args() | ||
| args.out_dir.mkdir(parents=True, exist_ok=True) | ||
|
|
||
| print(f"Loading model {MODEL_ID}@{MODEL_REVISION} (f32)") |
There was a problem hiding this comment.
Suggested change
| print(f"Loading model {MODEL_ID}@{MODEL_REVISION} (f32)") | |
| print(f"Loading model {MODEL_ID}@{MODEL_REVISION} (f32)", file=sys.stderr) |
For scripts and binaries if its not an output and just logging info use stderr. Stdout is reserved for tool output.
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
2 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.
Adds an end-to-end CPU quality regression for
microsoft/resnet-50,Qwen/Qwen3-0.6B, andopenai/whisper-tiny.en. IREE compiles the committedmodel.mlir, fetches the externalized parameters from the Hugging Face Hub (pinned by revision), runs onllvm-cpu, and compares the output logits against the committed reference.