strisunshinewentingwang/pytorch

Pull Requests Commits

Remove unnecessary py::object copy in PyRRef ctor (#38402)

Shen Li committed 4 years ago

f99a693c

[ROCm] increase timeout, enable test_backend_group (#36166)

jeffdaily committed 4 years ago

54c16b44

Migrate erfc from TH to ATen (CUDA) (#38373)

cloudhan committed 4 years ago

8d94615c

relax MAX_JOBS restriction for ROCm builds (#38425)

jeffdaily committed 4 years ago

beedc654

Updating submodules

svcscm committed 4 years ago

b1d2c176

Clean up error handling in is_nonzero and where in TensorCompare.cpp (#38150)

xuhdev committed 4 years ago

336e1ec5

allow user passing relative paths in include_dirs within setuptools.setup (#38264)

lixinyu committed 4 years ago

5a979fcb

[quant][graphmode][refactor] insertDeQuantForAllUse (#38277)

jerryzh168 committed 4 years ago

ee8bf1c6

[quant][graphmode][refactor] Refactor propagateQuantizationOps (#38276)

jerryzh168 committed 4 years ago

eb66dd0b

[JIT] [Easy] Add location to implicit conversions (#38442)

Elias Ellison committed 4 years ago

8d883f5c

[quant][graphmode] Move leaky_relu to general value op map (#38166)

jerryzh168 committed 4 years ago

7ce733d2

[quant][graphmode] Move elu to general value ops map (#38165)

jerryzh168 committed 4 years ago

16696186

[quant][graphmode] Move hardtanh to general value ops map (#38164)

jerryzh168 committed 4 years ago

98d78a7f

[quant][graphmode] Move clamp to general value ops map (#38163)

jerryzh168 committed 4 years ago

1fde373f

[quant][graphmode] Move interpolate to general value ops (#38162)

jerryzh168 committed 4 years ago

e988b4fb

[torchbind] Better error message when missing init. (#37474)

suo committed 4 years ago

0d220ef3

[jit] move torchbind tests to separate file (#37473)

suo committed 4 years ago

2efa7e04

[quant][graphmode] Add quantizedconv1d to graphmode (#38341)

supriyar committed 4 years ago

7d7d7365

[quant] Add quantized::conv1d op benchmarck (#38332)

supriyar committed 4 years ago

ae11718c

[quant] Add support for Quantized Conv1d and ConvRELU1d (#38283)

supriyar committed 4 years ago

f6626aaf

[quant] Add support for quantized::conv1d operator (#38248)

supriyar committed 4 years ago

2d221df5

Added autograd tests, disabled jit autograd tests for complex and added a separate list for tests for complex dtype only (#38399)

anjali411 committed 4 years ago

1676c7d6

improve some reporting for fakelowp tests (#38428)

Hector Yuen committed 4 years ago

53439be6

Move Cuda Abs kernel to its own file. (#38274)

gchanan committed 4 years ago

dac9b618

speed up hardswish and hardsigmoid tests (#38256)

vkuzo committed 4 years ago

ff76de8a

Use GIL to guard decref of jit::toPyObj return value in processRpc (#38376)

Shen Li committed 4 years ago

afa4dbd7

Update Cpp, rpc docs and Libraries section to match 1.5 (#38350)

Jessica Lin committed 4 years ago

33977ca7

[future] Make new IValue future constValue semantics match torch::utils counterpart (#38355)

jjlilley committed 4 years ago

328dd9e5

[quant][graphmode][refactor] Factor out common parts of general value ops (#38161)

jerryzh168 committed 4 years ago

b668bbc4

[TensorExpr] TensorExprKernel: don't do any compilation or lowering in run(). (#37948)

Mikhail Zolotukhin committed 4 years ago

6e13146d

Newer Older