Skip to content

[Tests] Increase maximum quantization error #1245

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.

Sign up for GitHub

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

Merged
kylesayrs merged 3 commits into from
Mar 13, 2025
Merged

[Tests] Increase maximum quantization error #1245

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
70aecd5
increase max quantization error
kylesayrs Mar 12, 2025
50dc250
fix typo
kylesayrs Mar 12, 2025
c68a28b
Merge branch 'main' into kylesayrs/reduce-noisy-test
kylesayrs Mar 13, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions tests/llmcompressor/transformers/sparsification/test_compress_tensor_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -430,8 +430,8 @@ def test_compressor_stacking(model_stub, recipe, sparse_format, quant_format, tm
if key.endswith("weight") and quant_format != "dense":
# we don't expect an exact match for compressed
diff = torch.abs(dense_tensor - reconstructed_tensor)
# max diff value found empirically
assert not torch.any(diff > 0.022), f"Max diff: {torch.max(diff)}"
# maximum quantization error as a result of compression is ~0.025
assert not torch.any(diff > 0.025), f"Max diff: {torch.max(diff)}"
else:
assert torch.equal(dense_tensor, reconstructed_tensor)
shutil.rmtree(tmp_path)
Expand Down