Supriya Rao
9cacbe29e5
[quant] Add reduce_range argument for qlinear_dynamic ( #39041 )
...
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/39041
reduce_range option restricts the activation tensor to 7 bits instead of 8.
This is necessary to enable per channel quant for RNNs and LSTMs
Test Plan:
python test/test_quantization.py TestDynamicQuantizedLinear
Imported from OSS
Reviewed By: akinh
Differential Revision: D21769691
fbshipit-source-id: ef0e9873367f3c1b34091b0b3af788233ef60c6c
2020-05-29 18:19:36 -07:00
Presley Graham
ff2e29144c
Refactor backward compatibility tests to use override_qengines decorator ( #38838 )
...
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/38838
Test Plan: Imported from OSS
Differential Revision: D21676032
Pulled By: durumu
fbshipit-source-id: 5cbe56e0d72d322f540bccffb60bcdbb15385ee8
2020-05-27 15:37:47 -07:00
Supriya Rao
138476389e
[quant] Disable qnnpack test when TSAN is enabled ( #38153 )
...
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/38153
Test fails when opt-tsan is enabled
Test Plan: buck test mode/opt-tsan //caffe2/test:quantization -- 'test_single_linear_dynamic \(quantization\.test_quantize\.TestGraphModePostTrainingStatic\)' --run-disabled
Reviewed By: vkuzo
Differential Revision: D21482799
fbshipit-source-id: fe6d1d84f525387081fabb90ce876c7c7dafd081
2020-05-08 16:52:36 -07:00
Supriya Rao
76c964dfb0
Reland [quant][tests] Enable tests to run on all qengine backends ( #37943 )
...
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/37943
Refactor tests to use supported_qengines
Test Plan:
python test/test_quantization.py
Imported from OSS
Differential Revision: D21435514
fbshipit-source-id: 8004ef2535e1cc65036f331c00af27ded1c04a6b
2020-05-06 22:38:50 -07:00
Supriya Rao
b33b46a950
[quant] Enable qnnpack tests for test_quantize and test_numeric_suite ( #37351 )
...
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/37351
Test Plan:
python test/test_quantization.py PostTrainingStaticQuant
Imported from OSS
Differential Revision: D21293704
fbshipit-source-id: 621f3ac60315b61f99b9b41da691ac3473e974cc
2020-04-29 19:28:22 -07:00
Mike Ruberry
dcd8a1b399
Revert D21286660: [quant] Generalizing _calculate_dynamic_qparams in quantized test
...
Test Plan: revert-hammer
Differential Revision:
D21286660
Original commit changeset: 98d90cdb34ac
fbshipit-source-id: a4194193c9aa53fb2dc9bbc04fde9c2925aa378f
2020-04-28 18:01:44 -07:00
Zafar Takhirov
239ce75a74
[quant] Generalizing _calculate_dynamic_qparams in quantized test ( #37451 )
...
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/37451
Test Plan: Imported from OSS
Differential Revision: D21286660
Pulled By: z-a-f
fbshipit-source-id: 98d90cdb34ac3d0ef33f7ebe1c9f32001d4e80b6
2020-04-28 16:55:51 -07:00
Zafar Takhirov
f463586739
Revert D20984966: [quant] Generalizing _calculate_dynamic_qparams in quantized test
...
Test Plan: revert-hammer
Differential Revision:
D20984966
Original commit changeset: 17437297adae
fbshipit-source-id: 30b9f7a2b2a772b2bf1c4b81cf99bddf37d4b179
2020-04-27 14:36:44 -07:00
Zafar
205c6ffbc5
[quant] Generalizing _calculate_dynamic_qparams in quantized test ( #36449 )
...
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/36449
Test Plan: Imported from OSS
Differential Revision: D20984966
Pulled By: z-a-f
fbshipit-source-id: 17437297adae813bc5c6fa43c6c7514f72ce2f6c
2020-04-25 17:06:40 -07:00
Supriya Rao
169541871a
Add operator support for dynamic quant on mobile ( #32479 )
...
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/32479
Run dynamic quantization on mobile (similar to FBGEMM). Currently only implemented on linear operator
Test Plan:
python test/test_quantized.py TestDynamicQuantizedLinear.test_qlinear
Imported from OSS
Differential Revision: D19542980
fbshipit-source-id: c9f6e5e8ded4d62ae0f2ed99e478c8307dde22ed
2020-01-24 17:51:54 -08:00
Pritam Damania
f050b16dd9
Move pytorch distributed tests to separate folder for contbuild. ( #30445 )
...
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/30445
Create distributed and rpc directories under caffe/test for better management
of unit tests.
Differential Revision: D18702786
fbshipit-source-id: e9daeed0cfb846ef68806f6decfcb57c0e0e3606
2020-01-22 21:16:59 -08:00
