strisunshinewentingwang/pytorch

Pull Requests Commits

[caffe2] Fix shape inference for Softmax (#53132)

khabinov committed 3 years ago

fdd074e8

Enable Kineto in CPU builds (#53174)

Ilia Cherniavskii committed 3 years ago

795ed5ca

[PyTorch Mobile] Fix case when error messages are stripped, and stack value isn't popped off in lite-interpreter (#53201)

dhruvbird committed 3 years ago

17495e03

Revert D26819810: Revert D26815021: Revert D26744062: Add assert_async

ezyang committed 3 years ago

1accffe4

Update foreach APIs to use scalar lists (#51893)

izdeby committed 3 years ago

110a17a4

[Static Runtime] remove redundant gather_ranges when fusing (#53323)

Marat Subkhankulov committed 3 years ago

47dbdfcf

[torch.futures] Add note about error handling for non-chained futures. (#53212)

rohan-varma committed 3 years ago

97d4ed3d

[Static Runtime] Remove dead code in MemoryPlanner and rename unmanaged_value_set to unmanaged_ivalue_set

Hao Lu committed 3 years ago

ac668c55

[static runtime] aten::to copy out variant (#52343)

ajyu committed 3 years ago

36180c13

make torch.load() aware of import path changes: torch.tensor -> torch._tensor (#53139)

bdhirsh committed 3 years ago

18277137

Remove MNIST for XLA (#53274)

cpuhrsch committed 3 years ago

a558d362

Fix gradfn attr bindings when saved variable is of an output (#53205)

soulitzer committed 3 years ago

a3c3141d

[PyTorch] Reduce size of register_symbols.cpp (#53278)

swolchok committed 3 years ago

6db2f012

Skip some nodes during discovery using sequence number (#52180)

soulitzer committed 3 years ago

4739d15a

Support submodule manipulation in GraphModule (#52358)

Ansley Ussery committed 3 years ago

85109ce4

Leak autograd threads after wait limit (#53170)

malfet committed 3 years ago

72ec7183

Update CODEOWNERS to be tagged as reviewer (#53277)

H-Huang committed 3 years ago

51718c2f

fix tensorpipe_agent linked even when USE_TENSORPIPE is turned off (#53281)

Rong Rong (AI Infra) committed 3 years ago

b0aa03b7

Edit SiLU documentation (#53239)

veeara282 committed 3 years ago

b4395b04

Parametrization Functionality (#33344)

lezcano committed 3 years ago

7aeee284

[PyTorch] Don't inline Dispatcher::call on mobile (#53197)

swolchok committed 3 years ago

3826a07a

Declare NamedTuple at top level (#53273)

ezyang committed 3 years ago

8c54cd7f

Revert D26815021: Revert D26744062: Add assert_async

ezyang committed 3 years ago

9e5e5a7d

Context manager for hiding source ranges (#53188)

jbschlosser committed 3 years ago

6557ea05

Optimize module path finding (#52990)

Ansley Ussery committed 3 years ago

6dce0cd0

Enabled amin & amax for float16 & bfloat16 (#52579)

imaginary-person committed 3 years ago

e698a634

Minor cleanup of interpolate microbenchmark

NicolasHug committed 3 years ago

5095332a

Revert D26744062: Add assert_async

Mike Ruberry committed 3 years ago

b8644577

[ROCm] Enable test in test_linalg.py, test_optim.py and test_vmap.py … (#52818)

KyleCZH committed 3 years ago

bf5e5bf9

[special] add `torch.special` namespace (#52296)

kshitij12345 committed 3 years ago

c4c77e20

Newer Older