strisunshinewentingwang/pytorch

Pull Requests Commits

Add undefined tensor gradient support to all backward functions (#39400)

kurtamohler committed 4 years ago

bba30d1b

.circleci: Move ecr gc build job to ecr gc workflow (#38523)

seemethere committed 4 years ago

8251f187

Fast tanh for the LLVM backend. (#39528)

zheng-xq committed 4 years ago

83dd5663

c10/util/complex_math.h and c10/util/complex_utils.h should not be individually included (#39276)

xuhdev committed 4 years ago

df2d1972

Cleanup rref_impl (#39530)

rohan-varma committed 4 years ago

397b24bb

[predictor] move fblearner/predictor to platform009

HC Zhu committed 4 years ago

a2125135

[ROCm] Enable AMP autocast tests on ROCm (#39616)

rohithkrn committed 4 years ago

ab6c447f

Revert D21930435: Revert D17923732: Optimize GroupNorm on CUDA

ezyang committed 4 years ago

cc2f7fa5

add dtype checks for scatter/gather family of functions. (#38646)

nikitaved committed 4 years ago

e4f9c74d

Remove duplicate 'with_gil' declaration. (#39540)

gchanan committed 4 years ago

e3e8f24c

Revert D17923732: Optimize GroupNorm on CUDA

ezyang committed 4 years ago

a83f7a1d

Add error message when negative stride is passed to as_strided (#39508)

wjgan7 committed 4 years ago

e41fe608

add overload name for min/max with list input (#39614)

linbinyu committed 4 years ago

820e81ba

[futures] Add c++ ivalue::Future collectAll() helper (#39119)

jjlilley committed 4 years ago

b83fed8d

[quant] QNNPACK deconv kernel and tests (#36790)

Zafar Takhirov committed 4 years ago

172f3117

[jit] avoid pre-convert tensor to cpu in pickling (#38898)

wanchaol committed 4 years ago

6c56671f

[quant] QNNPACK deconvolution packing (#37405)

Zafar Takhirov committed 4 years ago

1db4a31d

Fix smoke test jobs (#39638)

peterjc123 committed 4 years ago

ee2bc13f

remove double registered ops (#39609)

linbinyu committed 4 years ago

b06b792b

remove duplicated op schema for aten::pow (#39606)

linbinyu committed 4 years ago

81776373

Optimize GroupNorm on CUDA (#28204)

xiaomengy committed 4 years ago

614dd032

instancenorm: static quant graph mode support (#39096)

vkuzo committed 4 years ago

ebdff07d

groupnorm: graph mode static quant support (#39095)

vkuzo committed 4 years ago

b443ca26

layernorm: eager mode qat support (#39094)

vkuzo committed 4 years ago

952deba8

instancenorm: eager mode QAT support (#39093)

vkuzo committed 4 years ago

b530176d

groupnorm: eager mode QAT support (#39092)

vkuzo committed 4 years ago

202625ba

instancenorm: eager static quant support (#39091)

vkuzo committed 4 years ago

21408742

groupnorm: eager static quant support (#39090)

vkuzo committed 4 years ago

f9b675f7

quant: clean up normalization channels_last handling (#37802)

vkuzo committed 4 years ago

26bc2727

[quant][graphmode] Dynamic quantInsert observers for module output (#39458)

supriyar committed 4 years ago

8a4597b8

Newer Older