strisunshinewentingwang/pytorch

Pull Requests Commits

Fix Broken Link in CONTRIBUTING.md (#41066)

dongreenberg committed 4 years ago

bacca663

Add PyTorch Glossary (#40639)

gmagogsfm committed 4 years ago

445128d0

add first implementation of swish (#41085)

Hector Yuen committed 4 years ago

bce75a25

use PYTORCH_ROCM_ARCH to set GLOO_ROCM_ARCH (#40170)

t-vi committed 4 years ago

a8bc7545

.circleci: Fix job-specs-custom docker tag (#41111)

seemethere committed 4 years ago

054e5d89

add "aten::add.str" op and remove two duplicated ops

linbinyu committed 4 years ago

cc29c192

bump docker version to more recent tag (#41105)

seemethere committed 4 years ago

a4fd4905

Add bfloat16 support for nccl path (#38515)

jithunnair-amd committed 4 years ago

eea53574

ROCm 3.5.1 image (#40385)

jeffdaily committed 4 years ago

38b465db

Add support for int[]? arguments in native_functions.yaml (#37174)

dreiss committed 4 years ago

5e03a1e9

In interpolate, inline the call to _interp_output_size (#37173)

dreiss committed 4 years ago

4dad829e

In interpolate, move exceptional cases to the bottom (#37172)

dreiss committed 4 years ago

3c1c74c3

In interpolate, use if instead of elif (#37171)

dreiss committed 4 years ago

8f0e2547

Expose certain methods in OpaqueTensorImpl. (#41060)

pritamdamania committed 4 years ago

93778f3b

Decouple DataParallel/DistributedDataParallel from CUDA (#38454)

chengjunlu committed 4 years ago

8d570bc7

Doc warnings (#41068)

mattip committed 4 years ago

75155df8

.circleci: Output binary sizes, store binaries (#41074)

seemethere committed 4 years ago

ff3ba25b

Insert parentheses around kernel name argument to hipLaunchKernelGGL (#41022)

jithunnair-amd committed 4 years ago

0e6b7502

Splitting embedding_bag to embedding_bag_forward_only and embedding_bag (#40557)

gmagogsfm committed 4 years ago

630e7ed9

Fix link to PyTorch organization (from Governance) (#40984)

mathemage committed 4 years ago

00ee54d2

Grammatically updated the tech docs (#41031)

lento committed 4 years ago

452d5e19

If ninja is being used, force build_ext to run. (#40837)

ezyang committed 4 years ago

22c7d183

Fix several quantization documentation typos (#40567)

Edward Leardi committed 4 years ago

733b8c23

Add option to warn if elements in a Compare table are suspect (#41011)

Taylor Robie committed 4 years ago

2d98f817

Revert D22396896: [pytorch][PR] run single-threaded gradgradcheck in test_nn

nairbv committed 4 years ago

a04af4dc

type annotations for dataloader, dataset, sampler (#39392)

Baranowski committed 4 years ago

0e09511a

Make `torch_cpu` compileable when `USE_TENSORPIPE` is not set. (#40846)

malfet committed 4 years ago

a6b703cc

Remove unused Logger in get_matching_activations (#41023)

Haixin Liu committed 4 years ago

12b5bdc6

Fix unordered-map-over-enum for GCC 5.4 (#41063)

malfet committed 4 years ago

4aa543ed

Fix CUDA jit codegen compilation with gcc-5.4 (#41055)

malfet committed 4 years ago

50df0975

Newer Older