strisunshinewentingwang/pytorch

Pull Requests Commits

Threading and CPU Inference note

Ilia Cherniavskii committed 5 years ago

41dfe720

Support pack_padded_sequence and pad_packed_sequence

wanchaol committed 5 years ago

f4eb93f7

support torch._C._get_tracing_state in script

wanchaol committed 5 years ago

c384fbf4

Specify onnxruntime version to install for CI tests (#23517)

BowenBao committed 5 years ago

e1f89859

support torch.as_tensor in script

wanchaol committed 5 years ago

c779eff5

CI: install clang-tidy (#23518)

ethomson committed 5 years ago

3a568c9a

Add sanity checks for NCCL detection.

xuhdev committed 5 years ago

a8edc2b5

avoid Include the same header file twice (#23418)

dongfangduoshou123 committed 5 years ago

9219a37c

fix fbcode weak ordering (#23511)

Elias Ellison committed 5 years ago

dec4eaca

Fix TestCuda.test_events_wait (#23520)

yxsamliu committed 5 years ago

0c9979dd

Add multiprocessing_context= argument to DataLoader (#22990)

ssnl committed 5 years ago

e982e46d

Untap caskroom/homebrew-cask in attempt to unbreak OS X builds.

ezyang committed 5 years ago

56664c2c

add sorting policy to ChunkDataset (#23053)

xzhu1900 committed 5 years ago

31f19280

add note to Contribution Guide around recently released research (#23513)

soumith committed 5 years ago

a356276d

Fix distributions.Categorical.sample bug from .view() (#23328)

fritzo committed 5 years ago

06762b47

fixes #20178 (#23297)

shihongzhi committed 5 years ago

be644d82

Make randperm works properly on non-contiguous tensors. (#23043)

xuhdev committed 5 years ago

236149ed

fix an over-indented return in trace_module

Krovatkin committed 5 years ago

d6ff78fd

Implement named inference rule for mul

zou3519 committed 5 years ago

505fa83b

Try another version of apt/dpkg killing.

ezyang committed 5 years ago

d3fcb4cc

Remove two CMAKE_ build options from additional_options. (#23451)

xuhdev committed 5 years ago

8ada7c99

Whether MKLDNN should be built under native arch should respect USE_NATIVE_ARCH (#23445)

xuhdev committed 5 years ago

09ba4df0

Remove redundant MSVC_Z7_OVERRIDE processing and combine "/EHa" flag setup (#23455)

xuhdev committed 5 years ago

b335f391

Fix build issue. CUDA may be installed in `$CUDA_HOME/lib` on macOS. (#23491)

rgommers committed 5 years ago

81e46d4f

Let set_rng_state and get_rng_state accept string parameter (#23448)

xuhdev committed 5 years ago

97f129bf

Update PyTorch Docker image to 323. (#23389)

ezyang committed 5 years ago

7a820662

fixing documentation, issue 22697 (#23268)

Mingbo Wan committed 5 years ago

f546a3b8

Sync worker requirement mismatches

alisdair04 committed 5 years ago

19858f7c

Remove unused cuBLAS driver functions for getrs (#23375)

vishwakftw committed 5 years ago

91d28026

Add some compiler flags for building cpp extensions on Windows (#23472)

peterjc123 committed 5 years ago

54c28086

Newer Older