strisunshinewentingwang/pytorch

Pull Requests Commits

Store bias in PackedLinearWeight struct in fbgemm (#25428)

supriyar committed 5 years ago

9d2d31e6

fix OSS mobile CI (#25755)

ljk53 committed 5 years ago

4c7189d0

Use whitelist instead of blacklist for USE_DISTRIBUTED (#25759)

pietern committed 5 years ago

3e843115

remove tools/setup_helpers/cudnn.py (#25482)

xuhdev committed 5 years ago

66ac6698

Enable loading int8 prepacked models in PredictorContainer

csummersea committed 5 years ago

d95763b4

Do not pass down USE_GLOO_IBVERBS to CMake (#25720)

xuhdev committed 5 years ago

cc421106

Adds a -m flag to pytorch.distributed.launch (#24910)

Cameron Long committed 5 years ago

d47ced49

remove caffe2.pb.h dependency for embedding_lookup_idx.cc (#25670)

ljk53 committed 5 years ago

2bed2011

Expose an API to iterate all the registered operators (#23207)

houseroad committed 5 years ago

a6fb6e1f

Copy quantize routine to vec256 (#25685)

James Reed committed 5 years ago

21ba9b3c

Vectorized specialization of max_pool2d for channels-last layout (#25676)

James Reed committed 5 years ago

f7bcba33

Make tensor key in Dict works in serialization (#25442)

wanchaol committed 5 years ago

ed643382

Migrate digamma and polygamma from the TH to Aten (CUDA) (#25662)

ifedan committed 5 years ago

d939ee2d

Add CosineAnnealingWarmRestarts to optim documentation (#25421)

TortillasAlfred committed 5 years ago

38e47663

Improve handling of mixed-type tensor operations (#22273)

nairbv committed 5 years ago

88e4cee3

Enable jit fusion on ROCm (#22872)

iotamudelta committed 5 years ago

9c5a8997

add __getitem__ to class types (#25664)

Elias Ellison committed 5 years ago

82c8949a

Move most BUILD_NAMEDTENSOR macros out of header areas (#25721)

zou3519 committed 5 years ago

76bc44fb

Finish testing code examples in the docs (#25668)

davidriazati committed 5 years ago

0be29ee2

Enable two tests that were skipped b/c of rocThrust bugs fixed in ROCm 2.7

iotamudelta committed 5 years ago

c6dd4036

Cyclical learning rate multiplier: use fabs(base_lr) (#25628)

Swati Rallapalli committed 5 years ago

1559c644

Revert D17199043: [JIT] preserve ignored function return value type

suo committed 5 years ago

11eb8ac2

Align AliasInfo's operator<< with FunctionSchema (#23206)

houseroad committed 5 years ago

a294e157

Only default USE_DISTRIBUTED=True on Linux (#25725)

pietern committed 5 years ago

ce3b81fd

rocBLAS deprecated the last two parameters. (#25726)

iotamudelta committed 5 years ago

30aef56e

bring back skipped bitwise dispatch (#25689)

Ailing Zhang committed 5 years ago

bc2a37b2

Make SparseNormalize backwards compatible (#25660)

ffjiang committed 5 years ago

3be1745b

Adding RRef as return value for builtin operators (#25169)

mrshenli committed 5 years ago

197fd4f7

move USE_STATIC_DISPATCH from CI script to master cmake (#25696)

ljk53 committed 5 years ago

99b6472d

rename 'mobile' to 'static_dispatch' (#25695)

ljk53 committed 5 years ago

17e7079a

Newer Older