[quant][pt2e] Add reference representation for quantized add - relu #105707
Conversation
Summary: Implementing reference representation for quantized ops we decided in https://docs.google.com/document/d/17h-OEtD4o_hoVuPqUFsdm5uo7psiNMY8ThN03F9ZZwg/edit#heading=h.ov8z39149wy8 Test Plan: python test/test_quantization.py TestQuantizePT2E.test_representation_add_relu Although right now it is not really testing things since there is some problem with dynamo export Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/105707
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 8779f41: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
…dd - relu" Summary: Implementing reference representation for quantized ops we decided in https://docs.google.com/document/d/17h-OEtD4o_hoVuPqUFsdm5uo7psiNMY8ThN03F9ZZwg/edit#heading=h.ov8z39149wy8 Test Plan: python test/test_quantization.py TestQuantizePT2E.test_representation_add_relu Although right now it is not really testing things since there is some problem with dynamo export Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
|
|
||
| example_inputs = (torch.randn(1, 3, 3, 3), torch.randn(1, 3, 3, 3),) | ||
|
|
||
| self._test_representation( |
There was a problem hiding this comment.
this is not testing anything, is it?
There was a problem hiding this comment.
right now I just ran this locally and checked the graph result, and numerical result. we can't enable numerical result test right now due to timeout of export in sandcastle
I can add something for checking graphs, but for numerical tests I think we'll need the torchdynamo export to be fixed, I'll put some repro instructions next week
|
Also, we should really consider refactoring tests. Specifically representation tests can be separate. |
|
yeah I can do this after this week, there are a few things that I need to finish before the end of this week |
…dd - relu" Summary: Implementing reference representation for quantized ops we decided in https://docs.google.com/document/d/17h-OEtD4o_hoVuPqUFsdm5uo7psiNMY8ThN03F9ZZwg/edit#heading=h.ov8z39149wy8 Test Plan: python test/test_quantization.py TestQuantizePT2E.test_representation_add_relu Although right now it is not really testing things since there is some problem with dynamo export Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
…dd - relu" Summary: Implementing reference representation for quantized ops we decided in https://docs.google.com/document/d/17h-OEtD4o_hoVuPqUFsdm5uo7psiNMY8ThN03F9ZZwg/edit#heading=h.ov8z39149wy8 Test Plan: python test/test_quantization.py TestQuantizePT2E.test_representation_add_relu Although right now it is not really testing things since there is some problem with dynamo export Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Stack from ghstack (oldest at bottom):
Summary:
Implementing reference representation for quantized ops we decided in https://docs.google.com/document/d/17h-OEtD4o_hoVuPqUFsdm5uo7psiNMY8ThN03F9ZZwg/edit#heading=h.ov8z39149wy8
Test Plan:
python test/test_quantization.py TestQuantizePT2E.test_representation_add_relu
Although right now it is not really testing things since there is some problem with dynamo export
Reviewers:
Subscribers:
Tasks:
Tags: