strisunshinewentingwang/pytorch

Pull Requests Commits

clee2000 committed 2 years ago

342f7997

clee2000 committed 2 years ago

9e18ec49

only print if test step

clee2000 committed 2 years ago

5f72bb73

Sort: Use cub::WarpMergeSort for small sorts (32 < n <= 128) (#96223)

peterbell10 committed 2 years ago

91cce4c0

Add ModuleInfos for Adaptive{Max/Avg}Pool ops (#97291)

mikaylagawarecki committed 2 years ago

236bac81

Add gather to MTPG (#97555)

Rodrigo Kumpera committed 2 years ago

81770818

Use internal symbolizer for FBCODE (#97172)

zdevito committed 2 years ago

759e527e

run buildifier on the root BUILD.bazel file (#97611)

mikey dagitses committed 2 years ago

008be795

add device checks for sparse csr (#97520)

cybershiptrooper committed 2 years ago

bbc7c79b

[BE] Cleanup CMake flag suppressions (#97584)

malfet committed 2 years ago

96e3b3ac

Upload merge records to Rockset (#97471)

huydhn committed 2 years ago

345714e3

fix device type bug for custom device (#97213)

heidongxianhua committed 2 years ago

2ea09707

[BE] Update flake8-comprehensions to 3.11.1 (#97671)

Skylion007 committed 2 years ago

fcc312e9

Fix typo in throughput_benchmark. (#97619)

huihoaan committed 2 years ago

4d261137

[CI] Reduce perf nightly run frequency and bump up its timeout limit (#97682)

desertfire committed 2 years ago

97711ac6

Specify the head branch when upload perf stats to Rockset (#97643)

huydhn committed 2 years ago

6db196b7

Resubmit _int_mm (#96685)

cpuhrsch committed 2 years ago

9d37cefc

Remove hacky python dispatcher fallthrough (#96635)

angelayi committed 2 years ago

5f88d861

Dynamo size dim kwargs (#97450)

nima10khodaveisi committed 2 years ago

a6bc1f3a

[cpp_extension.py] fix bogus `_check_cuda_version` (#97602)

stas00 committed 2 years ago

8275e5d2

do not insert to_dtype for memory copy only buffers (#97147)

zhuhaozhe committed 2 years ago

a1ada050

Add support for copysign operator in functorch (#96018)

lithomas1 committed 2 years ago

e1f153f3

Remove unnecessary retain_grad call from gradcheck (#96923)

soulitzer committed 2 years ago

d0abc314

Modify all calls to checkpoint pass use_reentrant explicitly (#97376)

soulitzer committed 2 years ago

51c3fd39

Split rnn primitive for inference and training (#96736)

yanbing-j committed 2 years ago

38da54e9

[Dynamo] Unspec int list if enabling dynamic_shapes (#97557)

yanboliang committed 2 years ago

e3df6a7c

Specify file encoding in test_torch.py (#97628)

malfet committed 2 years ago

542fb0b1

[BE] Use nested namespaces in sparse (#97581)

malfet committed 2 years ago

b73e8cd4

add -std=c++17 to windows cuda compilations (#97515)

mikey dagitses committed 2 years ago

461f088c

[BE] Apply ufmt to run_test and GitHub Python util scripts (#97588)

huydhn committed 2 years ago

4c0dce50

Older