strisunshinewentingwang/pytorch

Pull Requests Commits

Gradle tasks for publishing to bintray, jcenter, mavencentral etc. (#25351)

IvanKobzarev committed 5 years ago

6e4eeb1d

Vectorized quantized relu/relu6 (#25496)

James Reed committed 5 years ago

a27fdfd3

remove PYTHON_VERSION (#25494)

suo committed 5 years ago

5455ba63

Manually implement `is_zipfile` (#25279)

davidriazati committed 5 years ago

7a921ba1

Minor fixes in per channel support for qconv kernel (#25182)

dskhudia committed 5 years ago

fcab254d

Revert "Memory layout for pooling ops" (#25495)

James Reed committed 5 years ago

0f928dc0

Disable Int8Transpose test

dreiss committed 5 years ago

77e8dba6

update speed benchmark binary to work in USE_STATIC_DISPATCH mode (#25449)

ljk53 committed 5 years ago

890d0f88

add speed benchmark binary for torch jit (#25486)

ljk53 committed 5 years ago

a7792635

turn off BUILD_BINARY for android CI jobs (#25485)

ljk53 committed 5 years ago

7ec6b74a

Fix implicit fallthrough warnings in FeatureLPPooling.cu (#25451)

rgommers committed 5 years ago

6d355799

Fix windows build error when TBB enabled and Windows SDK installed (#25398)

cloudhan committed 5 years ago

861194e3

Remove BUILD_ATEN_ONLY build option (#24441)

xuhdev committed 5 years ago

03f67e4b

Delete a few cases where we directly use Backend/TensorTypeId. (#25467)

ezyang committed 5 years ago

9bdcc499

Kill non-shared cwrap tools. (#25358)

gchanan committed 5 years ago

d159104d

Update derivatives.yaml docs to refer to Declarations.yaml rather than Declarations.cwrap. (#25357)

gchanan committed 5 years ago

28d4e2e9

add tuple keyword (#25474)

Elias Ellison committed 5 years ago

d2a8435c

Get rid of extract_cwarp. (#25356)

gchanan committed 5 years ago

7e61136c

Parameterize CircleCI config (#25446)

suo committed 5 years ago

fea4225b

Get rid of more unused plugins.

gchanan committed 5 years ago

fe055f2d

Get rid of torch._thnn. (#25354)

gchanan committed 5 years ago

a4fad42a

torch/jit/passes/quantization.{h,cpp} and torch/jit/init.cpp (#25403)

jerryzh168 committed 5 years ago

be0f8037

change shape for conv and unary ops (#25477)

hl475 committed 5 years ago

9d89c9a3

Migrate clamp and clamp_ from the TH to Aten (CPU) (#25290)

ifedan committed 5 years ago

1a92b225

cuda devices should have same dtype (#25470)

jfc4050 committed 5 years ago

e26305ed

Torch.flatten() returns a 1-dim tensor on a 0-dim tensor (#25406)

CamiWilliams committed 5 years ago

329757a9

yaroslavvb committed 5 years ago

4fb28e5d

Stop doing nn wrap. (#25353)

gchanan committed 5 years ago

25e6a52e

Stop initializing THNN backend. (#25352)

gchanan committed 5 years ago

716815e3

Compare shapes of outputs and grad_outputs in autograd.grad (#25349)

CamiWilliams committed 5 years ago

05bf74a8

Newer Older