Conversation
|
🧹 |
…ss 1 directory (#370) Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Fabien Casenave <[email protected]>
|
@fbordeu can you use draft mode while you still work on the PR ? Here, tests fails and the PR is in Open mode (it's for use to have the signal to start the review :-) ) |
fbordeu
force-pushed
the
BasicCleaning
branch
2 times, most recently
from
4f972ad to
ae1ca7b
Compare
|
|
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
| prediction[count][fn] = op.dot(sample_pred.get_field(fn)) | ||
| for sn in out_scalars_names: | ||
| prediction[count][sn] = sample_pred.get_scalar(sn) | ||
| prediction[count][sn] = sample_pred.get_global(sn) |
There was a problem hiding this comment.
I saw you changed the benchmarks, but the version of PLAID is fixed for theses benchmarks, and documented for each one. We shoud keep the original synthax
There was a problem hiding this comment.
ok, you want to keep get_scalar? the old implementation is just a redirection to get_global.
self.features.get_global(name)
and a global can be a numpy array; not a scalar. I think get_global is more appropriate.
There was a problem hiding this comment.
I'm not sure what was the exact behavior for each version tagged for each benchmark, so we're better off keeping the original code
|
|
5 participants
Thanks for contributing! Please make sure your PR title and content follow the guidelines.
Leave this checklist below in your PR description and tick the corresponding boxes.
Checklist
✅ PR Title Format
Your PR title must start with one of the following emojis to indicate the type of change:
Example:
🎉 Add MMD-based split method
📋 Description (optional)
Please explain what your PR does:
🔗 Related issues (optional)
Closes #___