Repository navigation
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/25178
Note: Links to docs will display an error until the docs builds have been completed. ❌ 3 Cancelled JobsAs of commit 93263ca with merge base 351e938 ( CANCELLED JOBS - The following jobs were cancelled. Please retry:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This PR needs a
|
|
@claude /pr-review Please review the current revision ( |
|
Claude finished @rascani's task in 20m 40s —— View job I found no correctness, quantization, or memory-safety bugs in the linear + ReLU/HardTanh fusion at Detailed reviewPR Review: #25178SummaryThe PR groups single-consumer I found no correctness, numerical, or memory-safety defects in the quantizer grouping, Code Quality
Testing
RecommendationApprove. Neither finding blocks the PR. Both are small hardening and test-strength follow-ups. Review steps
Not done: no Python or C++ tests were run, because running Python needs approval this job does not have. |
6267fb6 to
fe3fa73
Compare
fe3fa73 to
07491f0
Compare
07491f0 to
b659712
Compare
418b7c5 to
14266b4
Compare
ac600fd to
b722d7b
Compare
b722d7b to
56c4efc
Compare
Quantize single-consumer linear followed by ReLU or HardTanh, including ReLU6, at the pattern boundaries. Fold the final output qparams and clamp bounds into the existing ESP-NN fully connected call, preserving per-channel weights and derived int32 bias. Forked and separately quantized activations retain their original quantization boundaries. Read the quantizer and activation fold before the linear lowering and schema/wrapper updates. Shared dialect and S3 cases cover bias/no bias, forks, saturation, vector and higher-rank inputs, and narrower clamp grids with asymmetric, positive-only and negative-only HardTanh bounds. Authored with OpenAI Codex.
Fuse single-consumer linear followed by ReLU or HardTanh, including ReLU6, into the ESP-NN fully connected call. Quantization observes the pattern boundaries, and
FoldActivationstransfers the final output qparams and clamp bounds while preserving per-channel weights and derived int32 bias. Forked and separately quantized activations retain their boundaries.Review quantization and activation folding before the linear lowering and schema/wrapper updates. Shared dialect and S3 cases cover bias/no bias, forks, saturation, vector and higher-rank inputs, and narrow asymmetric, positive-only, and negative-only clamp grids.
Validation: linear-fusion and quantizer tests pass within the revised stack's 978-test Python suite. Scoped lint passed.
Authored with OpenAI Codex.