-
Notifications
You must be signed in to change notification settings - Fork 468
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
Feat: Implement ONNX RandomUniform + RandomNormal in burn-import #1806
Feat: Implement ONNX RandomUniform + RandomNormal in burn-import #1806
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1806 +/- ##
==========================================
- Coverage 86.46% 86.41% -0.05%
==========================================
Files 746 750 +4
Lines 86297 87092 +795
==========================================
+ Hits 74616 75263 +647
- Misses 11681 11829 +148 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice work! Some minor comments/changes but overall looks good 👍
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for contributing! 🙏
LGTM. One small typo to fix but I'll approve in advance.
🚀 🚀 |
Checklist
run-checks all
script has been executed (unrelated tests fail for me, see Burn-WGPU tests fail on Windows with Radeon 6950 #1805, but the ones for the import succeed)Related Issues/PRs
Implements a part of #1714
Changes
Adds a Node for ONNX RandomUniform to burn-import.
It supports all of the operation's attributes except
seed
, since I've found no way of providing a seed value to the Tensor (but since the specific Pseudorandom Generator isn't given, models cannot really rely on using fixed seeds anyways, so the impact is hopefully negligible)Testing
Created a testcase along the instructions of the contributor book. Since I've found no way of easily fixing the random seed in the testcase, it only checks the resulting tensor dimensions, not actual golden values.