strisunshinewentingwang/pytorch

Pull Requests Commits

Deprecates calling linspace and logspace without setting steps explicitly (#43860)

Mike Ruberry committed 4 years ago

7e91728f

Revert D23584071: [dper3] Create dper LearningRate low-level module

ngimel committed 4 years ago

e703c179

[dper3] Create dper LearningRate low-level module

Brandon Lin committed 4 years ago

a309355b

fuse layernorm + quantize (#44232)

Hector Yuen committed 4 years ago

0743d013

Add SNPE deps for caffe2 benchmark android binary

agarla committed 4 years ago

6f2c3c39

[ROCm] remove thrust workaround in ScanKernels (#44553)

jeffdaily committed 4 years ago

05c1f1d9

Cleanup workarounds for compiler bug of ROCm (#44579)

zasdfgbnm committed 4 years ago

d191caa3

fix dangling ptr in embedding_bag (#44571)

lixinyu committed 4 years ago

8641b551

Pass the input tensor vector by const reference. (#44340)

Yi Wang committed 4 years ago

82b44779

Move the inline implementations of GradBucket class to the header. (#44339)

Yi Wang committed 4 years ago

ab5fee27

[JIT] dont optimize device dtype on inline (#43363)

Elias Ellison committed 4 years ago

1f0dcf39

[TensorExpr] Do not inline autodiff graphs if they contain prim::TypeCheck nodes. (#44564)

Mikhail Zolotukhin committed 4 years ago

d729e296

[NNC] Cuda Codegen - mask loops bound to block/thread dimensions (#44325)

nickgg committed 4 years ago

64b4307d

Compile less legacy code when BUILD_CAFFE2 is set to False (take 2) (#44453)

malfet committed 4 years ago

2ae74c06

handle missing NEON vst1_*_x2 intrinsics (#44198) (#44199)

hrw committed 4 years ago

566b8d06

Change code coverage option name (#43999)

scintiller committed 4 years ago

db24c5c5

[quant][graphmode][fx][fix] Remove qconfig in convert (#44526)

jerryzh168 committed 4 years ago

b6f0ea0c

[fix] ReduceOps throw error if dim is repeated (#44281)

kshitij12345 committed 4 years ago

42f9f2f3

add `lcov` to oss for beautiful html report (#44568)

scintiller committed 4 years ago

f3a79b88

Filter default tests for `clang` coverage in oss

scintiller committed 4 years ago

c2b40b05

[quant][graphmode][fx][fix] Support None qconfig in convert (#44524)

jerryzh168 committed 4 years ago

a82ea6a9

removing conv filters from conv pattern matching (#44512)

z-a-f committed 4 years ago

1fb58830

Add iterator like functionality for DispatchKeySet (#44066)

linux-jedi committed 4 years ago

dd4bbe1a

Batched grad support for: slice, select, diagonal (#44505)

zou3519 committed 4 years ago

e2bb34e8

Add some batched gradient tests (#44494)

zou3519 committed 4 years ago

76324840

[rpc][jit] support remote call in TorchScript (#43046)

wanchaol committed 4 years ago

ab6126b5

[rpc][jit] support rpc_sync in TorchScript (#43043)

wanchaol committed 4 years ago

3e5df5f2

[rpc] rename some functions (#43042)

wanchaol committed 4 years ago

8bec7cfa

MinMax based observers: respect device affinity for state_dict (#44537)

vkuzo committed 4 years ago

70dfeb44

Simplify target handling in nn gradcheck. (#44507)

gchanan committed 4 years ago

192c4111

Newer Older