strisunshinewentingwang/pytorch

Pull Requests Commits

C++ parity, nn::CrossMapLRN2d (#29039)

lsrock1 committed 4 years ago

6389c187

Enable the intra-op parallelism for layer norm (#28464)

jianyuh committed 4 years ago

492764b1

Don't throw when type is used in TorchScript (#28053)

Elias Ellison committed 4 years ago

a5aeb374

Half test time, test_asymmetric_load_with_join, to avoid flakiness (#29139)

xush6528 committed 4 years ago

ac027d30

Cocoapods 1.3.1 release (#29240)

xta0 committed 4 years ago

ebf5dd44

Add cuda version for operators BatchSparseToDense and BatchDenseToSparse (#29166)

olittle committed 4 years ago

8a2dcff1

Generalized LU factorization (#28608)

pearu committed 4 years ago

fd4f22e4

submodule swapping via module interface (#28409)

wanchaol committed 4 years ago

9492994f

Revert D18299298: Migrate conv3d from TH to ATen (CPU)

zou3519 committed 4 years ago

f1c78492

README (#28533)

IvanKobzarev committed 4 years ago

eb418908

Revert D18309297: CPU-strided-complex support for ComplexFloat

ezyang committed 4 years ago

26f57cbe

assertEquals is deprecated, use assertEqual instead

zasdfgbnm committed 4 years ago

25e261d6

link to documentation for RNNBase.flatten_parameters() (#29196)

prasunanand committed 4 years ago

c99cdfeb

CPU-strided-complex support for ComplexFloat (#29133)

dylanbespalko committed 4 years ago

f32ab615

FindCUDA: Use find_program instead of find_path to find nvcc (#29160)

xuhdev committed 4 years ago

21d11e0b

Cleaned up func removed unused variable (#29179)

Zak Hassan committed 4 years ago

a02681f8

value assigned but never used in _recursive.py (#29181)

Zak Hassan committed 4 years ago

7434da2c

Support Conv+BatchNorm fusion for 1d/3d (#29113)

vadimkantorov committed 4 years ago

c6d908d4

Migrate conv3d from TH to ATen (CPU) (#29007)

andreaskoepf committed 4 years ago

546ae300

batch_norm_cpu_inference for channel last (#28982)

Xinyu Li committed 4 years ago

f2a35db2

support for cdist (#29129)

zou3519 committed 4 years ago

cb6d9dee

Add TensorNames::checkUnique, operator<< (TensorName) (#29124)

zou3519 committed 4 years ago

3233a058

Fix poisson_nll_loss with full option (#28637)

hi-ogawa committed 4 years ago

2c3c702d

Run clang-format for torch/distributed/rpc

pietern committed 4 years ago

49fba352

Rename PythonUDF{Call,Resp} (#27530)

pietern committed 4 years ago

6c391564

Scope pybind11 functions to torch.distributed.{autograd,rpc}

pietern committed 4 years ago

b4df4137

C++ API parity: MarginRankingLoss

pbelevich committed 4 years ago

69f845cb

Updating submodules

svcscm committed 4 years ago

0d056e75

use fbgemm's 3d group conv fast path (#29085)

jspark1105 committed 4 years ago

ca7d0803

Fix for torch.save not saving source files (#28965)

chauhang committed 4 years ago

9e314f55

Newer Older