strisunshinewentingwang/pytorch

Pull Requests Commits

PNP: early FX numeric suite tool to quantize each layer N times (#80521)

vkuzo committed 1 year ago

237316aa

make torch.histc ignore NaNs on CPU (#85870)

yuguo68 committed 1 year ago

b233d834

[Static Runtime] Block linalg_svdvals codegen & run codegen script (#85983)

Mike Iovine committed 1 year ago

ddec1eea

Fix doc of DDP (#86244) (#86256)

yhcharles committed 1 year ago

bebd1622

add myself for dynamic shapes PR review (#86292)

bdhirsh committed 1 year ago

020f2b2c

add nominal support for int32 indices in index/index_put ops (#86309)

ngimel committed 1 year ago

dc9c507d

Rename fromIntArrayRef to fromIntArrayRefSlow, audit call sites (#86235)

ezyang committed 1 year ago

e8b0bea6

Revert "Symintify getitem and add the required helper functions (#86207)"

pytorchmergebot committed 1 year ago

168ba066

Remove DataParallel remnants from DDP doc (#86221)

rohan-varma committed 1 year ago

be4e43c7

Mark ctc_loss with dynamic_output_shape (#86293)

SherlockNoMad committed 1 year ago

9e1a4312

Fix horribly double truncation bug in Scalar (#86304)

ezyang committed 1 year ago

0e5a27fb

[ao] fixing public v private for quantization_mappings.py (#86025)

HDCharles committed 1 year ago

73777d8a

[ao] fixing public v private for quantize_jit.py (#86024)

HDCharles committed 1 year ago

28a5cd94

Symintify getitem and add the required helper functions (#86207)

albanD committed 1 year ago

17addb30

Fix black binary again for debug python (#86275)

albanD committed 1 year ago

b8895df8

Revert "Revert "SymIntify cat and narrow (#86191)"" (#86289)

ezyang committed 1 year ago

e778fbf5

Install c10d headers with absolute path (#86257)

minsii committed 1 year ago

089a64e9

Fix ref / decomposition index_add (#86266)

lezcano committed 1 year ago

b67e0228

[ao] fixing public v private for quantize.py (#86023)

HDCharles committed 1 year ago

14db44ad

[ao] correctly set public v private for fake_quantize.py (#86022)

HDCharles committed 1 year ago

c21caff8

Optimize is_symbolic test and some refactor (#86230)

ezyang committed 1 year ago

3b1ec751

Log a new "timer expired" event to Scuba in file_based_local_timer (#85861)

bchen2020 committed 1 year ago

8c6d352b

Revert "SymIntify cat and narrow (#86191)"

pytorchmergebot committed 1 year ago

fc94a211

Improve make_tensor performance for float and complex types (#85473)

peterbell10 committed 1 year ago

3ec71fce

[torchdynamo hash update] update the pinned torchdynamo hash (#85774)

pytorchmergebot committed 1 year ago

7f607e8c

[MPS] Fix GELU for `torch.half` (#86218)

malfet committed 1 year ago

Verified 97d2e1df

SymIntify cat and narrow (#86191)

wconstab committed 1 year ago

63d8d4f6

Remove softmax from recomputable ops (#86268)

Chillee committed 1 year ago

0e03dc5f

Add refs for torch.unfold and a decomposition for its backward. (#85629)

lezcano committed 1 year ago

c6097688

[FSDP] Dequeue one instead of flush (#86165)

awgu committed 1 year ago

67eb2d59

Newer Older