strisunshinewentingwang/pytorch

Pull Requests Commits

[TensorExpr] Docs: checkin 'Conditionals in TE' doc. (#56949)

Mikhail Zolotukhin committed 3 years ago

8a0eb7fb

[static runtime] binding for aten::argmin_out (#56638)

ajyu committed 3 years ago

e909ad2d

ns for fx: additional bugfix for user defined functions (#56762)

vkuzo committed 3 years ago

9bd14da6

ns for fx: allow comparing int8 to int8 for functionals (#56742)

vkuzo committed 3 years ago

502c58ad

ns for fx: add option to skip matching classes and functions (#56493)

vkuzo committed 3 years ago

92c7aec5

ns for fx: support binary ops when adding unshadowed loggers for inputs (#56408)

vkuzo committed 3 years ago

c004346c

ns for fx: bug fix for shadowing fp16 emulation patterns (#56384)

vkuzo committed 3 years ago

f35540be

ns for fx: add fp16 function shadowing (#56311)

vkuzo committed 3 years ago

96a9eafc

ns for fx: allow user functions in shadowing (#56301)

vkuzo committed 3 years ago

19173509

ns for fx: move node I/O dtype mapping to be local instead of global (#56296)

vkuzo committed 3 years ago

93de8020

ns for fx: handling for user functions in weight and unshadowed act APIs (#56292)

vkuzo committed 3 years ago

8dbf6ae8

ns for fx: enable user defined functions for graph matching (#56283)

vkuzo committed 3 years ago

d405d41a

add deterministic path for index_copy_cpu (#56900)

yuguo68 committed 3 years ago

f5c24cc8

[static runtime] binding for aten::clamp_min_out (#56635)

ajyu committed 3 years ago

0888b872

[iOS GPU] Use thread buffer to store indices for transpose (#56706)

xta0 committed 3 years ago

d221be6f

Add reasons in TODO for the unblocked AVNTM -> InferenceMode cases. (#56823)

Ailing Zhang committed 3 years ago

16710e5d

[Static Runtime] Clean up op implementations (#56841)

Hao Lu committed 3 years ago

e810bed6

Added sm_75 to CUDA Arch List for Linux CI GPU builds (#56619)

mstfbl committed 3 years ago

9b46b6b3

Let RRef getValue() synchronize CUDA streams (#56895)

mrshenli committed 3 years ago

d1088de5

[caffe2] fix -Wrange-loop-construct

Igor Sugak committed 3 years ago

e1a7ec3c

add deterministic path for index_add_cuda (#56521)

yuguo68 committed 3 years ago

72c3ee07

.github: Add options to force unzip artifacts (#56929)

seemethere committed 3 years ago

cb1e7803

Clang format dist_utils.py and rpc/__init__.py (#56853)

Yi Wang committed 3 years ago

7989f2ac

[reland] Trigger azure pipeline for multi gpu tests (#56128)

skyline75489 committed 3 years ago

6155b0d9

fix bug in rocm device type (#56646)

Rong Rong (AI Infra) committed 3 years ago

2639c4e6

catch xml parser error during report test result phase in CI (#56864)

walterddr committed 3 years ago

2f598b53

fix ddp logging test (#56640)

zhaojuanmao committed 3 years ago

28a9483e

Add cuBLAS path for batched torch.geqrf (#56253)

IvanYashchuk committed 3 years ago

5b1f0ef6

Add cuSOLVER path for torch.geqrf (#56252)

IvanYashchuk committed 3 years ago

27a8ece8

Port CUDA torch.geqrf to ATen (#56251)

IvanYashchuk committed 3 years ago

f84f2063

Newer Older