soulitzer 958651327f Set default qengine to QNNPACK on ARM for quantization tests (#83097) ... The issue is that there is no FBGEMM or MKLDNN on ARM and QNNPACK is disabled by default for some reason. Not sure if this PR is the right approach, alternatives are probably: 1) keep these tests skipped 2) not disable QNNPACK by default on ARM 3) ? Pull Request resolved: https://github.com/pytorch/pytorch/pull/83097 Approved by: https://github.com/jerryzh168, https://github.com/albanD		2022-08-11 20:01:53 +00:00
..
__init__.py
test_equalize_fx.py	[Quant][fx] Hide equalization_config from prepare APIs (#80164)	2022-06-28 04:20:34 +00:00
test_model_report_fx.py	[ao] Create framework for ModelReport Qconfig Generation (#83091)	2022-08-11 00:11:50 +00:00
test_numeric_suite_fx.py	[Quant][fx][bc-breaking] Integrate BackendConfig with quantization flow (part 2) (#82557)	2022-08-08 18:55:50 +00:00
test_quantize_fx.py	Set default qengine to QNNPACK on ARM for quantization tests (#83097)	2022-08-11 20:01:53 +00:00
test_subgraph_rewriter.py