strisunshinewentingwang/pytorch

Pull Requests Commits

Change dper3 loss module to match dper2 (#28265)

Peiyao Zhou committed 5 years ago

46fefc98

torch.nn.functional.gumbel_softmax #27078 (#28121)

Naresh1318 committed 5 years ago

bd6f9e1d

Fix quantized avg_pool2d test to support non-zero padding (#28246)

supriyar committed 5 years ago

3629974c

Fix typo (#28281)

andreaskoepf committed 5 years ago

4b64ada5

String optimizations related to serialization. (#28230)

jjlilley committed 5 years ago

3d745508

String opts related to deserialization. (#28263)

jjlilley committed 5 years ago

ac61adb5

Implement lerp's derivative w.r.t. weight (#28219)

hi-ogawa committed 5 years ago

a1ac1508

Adding MSELoss, KLDivLoss and BCELoss to C++ front-end (#27156)

ShahriarSS committed 5 years ago

91a260ce

Disable blobs_queue_db_test in ROCm CI (#28268)

bddppq committed 5 years ago

9c41b61e

Clean up the stale item in bc white list (#28269)

houseroad committed 5 years ago

53d9456a

Minor: add static_assert to Pickler buffering. (#28114)

jjlilley committed 5 years ago

5c768ec3

In torch::save() avoid zip compressing small header records. (#28180)

jjlilley committed 5 years ago

d7ff34c0

Add tests for libtorch macOS binary (#25208)

Will Feng committed 5 years ago

5498a15d

Fix RNN nonlinearity (#28058)

davidriazati committed 5 years ago

2e7dd547

Disable c10d test_sync_params_with_buffers on ROCm (#28190)

bddppq committed 5 years ago

0b243e9c

cdist performance improvement for euclidean distance (#25799)

ifedan committed 5 years ago

12dde7f5

default caffe2_tvm_min_ops to 10 (#28250)

ajyu committed 5 years ago

7c1df06e

Add default arg to `prepare_qat` mapping. (#28193)

z-a-f committed 5 years ago

07b5666a

Address review comments on #28011.

Owen Anderson committed 5 years ago

7ebe8328

Export update for arange and _dim_arange (#26875)

neginraoof committed 5 years ago

95922c90

Changing observer name

z-a-f committed 5 years ago

a5ac7f63

Port of multi_margin_loss from TH to ATen (CPU) (#28062)

andreaskoepf committed 5 years ago

86e7e872

Add doc copy-edits from review (#26322)

davidriazati committed 5 years ago

618cb40e

change linear benchmark shapes (#28228)

mingzhe09088 committed 5 years ago

5c2bf8ab

Disable schema inference for unboxedOnly kernels (#27977)

smessmer committed 5 years ago

21c39979

Enables non-default CUDA stream in test_nn (#28192)

Mike Ruberry committed 5 years ago

8fff54ec

Add memory format support to typecasting shortcuts `byte`,`char`,`double`,`bool`,`half`,`int`,`long`,`short`,`float`,`bfloat16` (#27228)

VitalyFedyunin committed 5 years ago

951dd030

Add memory format support to typecasting shortcuts `byte`,`char`,`double`,`bool`,`half`,`int`,`long`,`short`,`float`,`bfloat16` (#27228)

VitalyFedyunin committed 5 years ago

15df3719

Fixing dispatch error in windows debug builds (#24360)

rje committed 5 years ago

c36552c4

out-variant for torch.batch_norm_elemt (#27621)

vadimkantorov committed 5 years ago

e1be08fc

Newer Older