[Bug Fix] Inverted bias flag in swap_conv2d_1x1_to_linear - #4618
[Bug Fix] Inverted bias flag in swap_conv2d_1x1_to_linear#4618gaurav-redhat wants to merge 1 commit into
Conversation
bias=(conv.bias is None) creates a bias when conv has none and skips it when conv has one. Fixed to bias=(conv.bias is not None). Also fixed -0 to 0 in PermuteSandwich.forward permute call.
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/4618
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
Hi @gaurav-redhat! Thank you for your pull request and welcome to our community. Action RequiredIn order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks! |
|
Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks! |
meta-cla
Bot
added
the
CLA Signed
|
Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks! |
1 participant
Summary
#4615
biasflag inreplace_conv2d_1x1insideswap_conv2d_1x1_to_linear.bias=(conv.bias is None)was creating a bias when the original conv had none. Changed tobias=(conv.bias is not None).-0typo to0inPermuteSandwich.forward.Impact
Low — the bug is self-healing because
lin.bias = conv.biason the next line overwrites the constructor value. The fix eliminates unnecessary Parameter allocation and reset_parameters() compute when converting bias-free convolutions.Test plan