strisunshinewentingwang/pytorch

Pull Requests Commits

Unify quantized conv and linear tests (#26992)

supriyar committed 4 years ago

b805b5da

Updating submodules

svcscm committed 4 years ago

38001fec

Move comment back to the class it describes (it accidentally drifted away).

Mikhail Zolotukhin committed 4 years ago

8e4031dd

Disable flaky distributed autograd test under spawn mode

mrshenli committed 4 years ago

e09c97b1

Enable tests (#27103)

jerryzh168 committed 4 years ago

98c02e6d

insert_prepack_unpack pattern (#27102)

jerryzh168 committed 4 years ago

14d29aee

API - add more passes to graph mode (#27093)

jerryzh168 committed 4 years ago

f742ceaa

Uninitialize the accumulation buffer to save some overhead (#27005)

jianyuh committed 4 years ago

6a096764

make rpc and dist-autograd multiprocess test to use both fork and spawn (#25656)

zhaojuanmao committed 4 years ago

1d2d59dd

Ensure that DDP wrapped module has parameters that require gradients (#25858)

danthe3rd committed 4 years ago

46539eee

Fixed seek offset size to 64bit. (#27125)

peterjc123 committed 4 years ago

ec07d144

Delete 5 THVector function declarations that have already been migrated to ATen.

xuhdev committed 4 years ago

48cd66cd

Bump gloo (#27133)

pietern committed 4 years ago

4086f743

Add C++ torch::nn::functional::pdist (#27122)

jon-tow committed 4 years ago

18eea826

Automatically select proper tqdm submodule (#27108)

ssnl committed 4 years ago

5ccecabb

Creates device generic cuDNN decorators (#26791)

Mike Ruberry committed 4 years ago

30997320

Move imports to after is_available in test_rpc to fix windows builds (#27126)

rohan-varma committed 4 years ago

85abfc19

Fuse module enhancements (#26457)

raghuramank100 committed 4 years ago

dddae3f8

Support layout() in script

James Reed committed 4 years ago

6a4ca9ab

Add C++ torch::nn::HingeEmbeddingLoss (#27101)

jon-tow committed 4 years ago

209dc4c4

Adds Device Generic Precision Tests to test_torch.py (#26762)

Mike Ruberry committed 4 years ago

ea414e49

Updating submodules

svcscm committed 4 years ago

1d4d6b6f

Add control for observers in Fake-quantize module (#27113)

raghuramank100 committed 4 years ago

9e3ba355

Support for add relu functional module (#26612)

raghuramank100 committed 4 years ago

bdcaf633

Cuts test_torch.py runtime in half by marking four tests as slow (#26789)

Mike Ruberry committed 4 years ago

ec7913af

make class types callable (#26743)

Elias Ellison committed 4 years ago

990f4ca7

Set number of threads for operator_benchmarks (#27010)

dskhudia committed 4 years ago

bf7ebc5a

Fix `toIValue` dict iteration (#26856)

davidriazati committed 4 years ago

5e7549c3

separate out rpc to sync and async apis (#26570)

rohan-varma committed 4 years ago

fb8fd6dc

Module method destroy

IvanKobzarev committed 4 years ago

3e20d9c0

Newer Older