strisunshinewentingwang/pytorch

Pull Requests Commits

Misc. FakeLowP OSS cleanup (#44331)

venkatacrc committed 4 years ago

364d03a6

BUG: make cholesky_solve_out do broadcast, error checking (#43137)

mattip committed 4 years ago

758c2b96

Use compile time cudnn version if linking with it statically (#44402)

malfet committed 4 years ago

683380fc

Fix frac in CUDA fuser (#44152)

bertmaher committed 4 years ago

6ec8fabc

Prevent the TE fuser from getting datatypes it can't handle (#44160)

bertmaher committed 4 years ago

350130a6

[te] Fix casting of unsigned char, and abs(int) (#44157)

bertmaher committed 4 years ago

960c088a

Skipping CUDA tests in ProcessGroupGloo and logs (#42488)

osalpekar committed 4 years ago

7c464eed

Autocast wrappers for RNN cell apis (#44296)

mcarilli committed 4 years ago

2a87742f

Moves some of TestTorchMathOps to OpInfos (#44277)

Mike Ruberry committed 4 years ago

a953a825

Disable a test (#44348)

Krovatkin committed 4 years ago

f044b17a

Don't use VCOMP if Intel OMP is used (#44280)

peterjc123 committed 4 years ago

cfd3620b

Remove pybind11 from required submodules (#44278)

Flamefire committed 4 years ago

d23f3170

Dump optimized graph when logging in already-optimized PE (#44315)

bertmaher committed 4 years ago

8acce550

Export Node::isBefore/isAfter for PythonAPI (#44162)

taewookoh committed 4 years ago

7a64b0c2

[Caffe2] Add RMSNormOp (#44338)

xiaomengy committed 4 years ago

135ebbde

Rename test_distributed to test_distributed_fork (#42932)

rohan-varma committed 4 years ago

106459ac

Enable test_distributed to work with spawn mode (#41769)

rohan-varma committed 4 years ago

b22abbe3

[quant] fill_ path for quantized tensors (#43303)

z-a-f committed 4 years ago

1d01fcdc

Resolve Autograd key for disable_variable_dispatch flag. (#44268)

Ailing Zhang committed 4 years ago

4aacfab2

Port nonzero cuda from THC to ATen (#44259)

ngimel committed 4 years ago

ecc6358d

[TensorExpr] Fuser: check node inputs' device before merging the node into a fusion group. (#44241)

Mikhail Zolotukhin committed 4 years ago

bd8e38cd

[quant] Move EmbeddingBag eager quantization to static (#44217)

supriyar committed 4 years ago

646ffd48

[quant] Add quantized Embedding module (#44208)

supriyar committed 4 years ago

57b87aaf

[quant] Support quantization of embedding lookup operators (#44207)

supriyar committed 4 years ago

6013a29f

[caffe2] fix wrong comment (#42735)

jspark1105 committed 4 years ago

f27be2f7

fix lint (#44346)

Elias Ellison committed 4 years ago

f9146b45

[quant][graphmode][fx][api] Call fuse in prepare (#43984)

jerryzh168 committed 4 years ago

6269b6e0

[NNC] fix support for FP16 in CudaCodgen (#44209)

nickgg committed 4 years ago

be94dba4

[quant][graphmode][fx] Support inplace option (#43983)

jerryzh168 committed 4 years ago

9f54bcc5

add rocm nightly build (#44250)

Rong Rong committed 4 years ago

0351d317

Newer Older