Riley Dulin 3be150653c [torch][ao] Add customizable loss function to NodeAccuracySummary (#136282) ... Summary: Add a customizable loss function callback to NodeAccuracySummary to allow users to pass in their own loss function. Also, fix some type errors and propagate better exception messages when unexpected tensor comparisons occur. Finally, enhance the robustness of `generate_numeric_debug_handle` in the case where it is called multiple times on the same model, by avoiding reuse of the same IDs. Test Plan: Added a test for this case in `test_numeric_debugger`. Reviewed By: jerryzh168 Differential Revision: D62898297 Pull Request resolved: https://github.com/pytorch/pytorch/pull/136282 Approved by: https://github.com/jerryzh168		2024-09-24 03:28:12 +00:00
..
test_duplicate_dq.py	Replace capture_pre_autograd_graph with export_for_training in torch tests (#135623)	2024-09-11 19:23:08 +00:00
test_graph_utils.py	Add None return type to init -- tests (#132352)	2024-08-01 15:44:51 +00:00
test_metadata_porting.py	[Dynamo] Trace torch function modes entered outside of torch.compile (#133137)	2024-09-14 18:52:22 +00:00
test_numeric_debugger.py	[torch][ao] Add customizable loss function to NodeAccuracySummary (#136282)	2024-09-24 03:28:12 +00:00
test_quantize_pt2e_qat.py	"Remove BLOCK_LIST" (#135729)	2024-09-12 01:22:06 +00:00
test_quantize_pt2e.py	Fix attr check for quantization spec (#135736)	2024-09-13 23:01:22 +00:00
test_representation.py	Add None return type to init -- tests (#132352)	2024-08-01 15:44:51 +00:00
test_x86inductor_quantizer.py	Replace capture_pre_autograd_graph with export_for_training in torch tests (#135623)	2024-09-11 19:23:08 +00:00
test_xnnpack_quantizer.py	[export][training ir migration] quantized_decomposed.quantize_per_tensor decomposition (#134525)	2024-09-06 07:06:06 +00:00