[ExecuTorch][WebGPU] Op-tests for native_group_norm#21174
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21174
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ❌ 77 New FailuresAs of commit 0de75be with merge base 266e0dc ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This PR needs a
|
psiddh
left a comment
There was a problem hiding this comment.
Approving full WebGPU stack
Stack from ghstack (oldest at bottom):
Adds the op-test suite for the
native_group_normop ported in the diff below, using the multi-output golden support to verify all 3 outputs.Key changes:
test/ops/test_group_norm.py—GroupNormModulereturns the fulltorch.native_group_norm(out, mean, rstd) tuple, so the multi-output golden path checks both the reduce (mean/rstd) and normalize (out) passes. +GroupNormTestdelegation smoke test.test/op_tests/cases.py—_group_norm_suite: c4_g2, c6_g3, c4_g1.Differential Revision: D112257626