strisunshinewentingwang/pytorch

Pull Requests Commits

Fix Binomimal overflow when logits is large (#20679)

fehiepsi committed 5 years ago

07197147

Re-sync with internal repository (#20702)

ezyang committed 5 years ago

Verified 9b1dbffb

Add get/set_num_interop_threads into torch.h include (#20659)

Ilia Cherniavskii committed 5 years ago

5835165c

better handling of getenv

Dmytro Dzhulgakov committed 5 years ago

45987293

Lightweight logging for once-only API usage

Dmytro Dzhulgakov committed 5 years ago

d3059b9c

separate option for FE_OVERFLOW (#20476)

jspark1105 committed 5 years ago

7b9ee598

Replace AT_ASSERT with TORCH_INTERNAL_ASSERT/TORCH_CHECK (#20668)

nishantpdce committed 5 years ago

dd050b7b

Support plot norm of specific embeddings of a LUT in diagnose_options (#19809)

Chenguang Xi committed 5 years ago

96a1f769

delete brodcasting ops from shape analysis resize aliasing (#20661)

wanchaol committed 5 years ago

23082574

De-deprecate parts of the legacy API (#20561)

smessmer committed 5 years ago

e7486947

Options based registration API (#20514)

smessmer committed 5 years ago

cb6be424

Add qint8 type (int8_t) (#19984)

jerryzh168 committed 5 years ago

85fad059

Add a pybind for Module::get_functions. (#20594)

Mikhail Zolotukhin committed 5 years ago

986c9eb5

Modify the Insert quant-dequant test cases to look for q-dq pattern (#20672)

nishantpdce committed 5 years ago

2af5911a

Some small performance fixes for c10 dispatcher (#20472)

Roy Li committed 5 years ago

e42665cf

Improve CPUAllocator OOM message (#20618)

Dmytro Dzhulgakov committed 5 years ago

d9dcfacd

Fix missing env for update_binary_size job

pjh5 committed 5 years ago

e79610c0

Misc error message improvements (#19369)

davidriazati committed 5 years ago

c267d0c8

SubgraphMatcher: add attributes support. (#20602)

Mikhail Zolotukhin committed 5 years ago

3be2f7c8

Use python type string for user facing error msgs (#20657)

Elias Ellison committed 5 years ago

0b9b929d

Add list(string) (#20617)

davidriazati committed 5 years ago

c819d767

Add `_enable_recursive_script` to try to script all Python functions (#19578)

davidriazati committed 5 years ago

a543586b

Add support for __getstate__/__setstate__ on module (#20242)

davidriazati committed 5 years ago

cd28ff53

Split out gpu/cpu targets based on gpu_library_targets (#20633)

lertoz committed 5 years ago

5f14ef8c

Remove unnecessary format literals from error message.

ezyang committed 5 years ago

79c5dc31

Remove TH/THC link for single matrix inverse (#20534)

vishwakftw committed 5 years ago

26dfeffa

Revert D15393514: [pytorch][PR] Refine CosineAnnealingWarmRestarts doc for issue #20028

ezyang committed 5 years ago

839a69f5

Add matmul optimization for the case A.ndim <= 2 && B.ndim >= 3 (#20448)

skrah committed 5 years ago

8c9f4c56

ir.cpp, module.cpp: clang-format. (#20592)

Mikhail Zolotukhin committed 5 years ago

a212a5b9

Don't split 256-bit AVX2 load/store intrinsics (#20609)

colesbury committed 5 years ago

b90790ab

Newer Older