strisunshinewentingwang/pytorch

Pull Requests Commits

Fix README for installation from source (#37301)

kventinel committed 4 years ago

5c628ddb

Let >> and << support half on CUDA (#37670)

xuhdev committed 4 years ago

3b97723f

graph mode: more in-place activation handling (#37771)

vkuzo committed 4 years ago

3673a724

graph mode: round out relu support (#37592)

vkuzo committed 4 years ago

b354700e

uninitialize output and bag_size in the fast path of EmbeddingBag to save overhead (#36681)

xing-liu committed 4 years ago

0b693e96

Migrate `erf` and `erf_` from the TH to Aten (CUDA) : Closes #24558 (#36724)

kshitij12345 committed 4 years ago

145560f4

[JIT] Fix GetAttr inconsistency (#37424)

Elias Ellison committed 4 years ago

23d0441d

Migrate clamp from the TH to Aten (CUDA) (#37646)

cloudhan committed 4 years ago

12e64916

[aten] Pass std::function<> to thread_pool by value, instead of const ref. (#37681)

jjlilley committed 4 years ago

468a9d44

Migrate CUDA unary complex kernel to c10::complex (#37647)

zasdfgbnm committed 4 years ago

d7ccb4b3

Enable test_distributed test test_backend_full_group (#37794)

jithunnair-amd committed 4 years ago

51c94442

Add vec256 for c10::complex (#37690)

zasdfgbnm committed 4 years ago

7c294489

Fix for hooks with no name (#37785)

albanD committed 4 years ago

6133be31

Migrate CUDA fill kernel to c10::complex (#37651)

zasdfgbnm committed 4 years ago

16c7907a

Revert D21369541: Make a separate cmake option for caffe2 tests

nairbv committed 4 years ago

d4edbbd3

[Tensorpipe/RPC] tensorpipe RPC agent (#35483)

jiayisuse committed 4 years ago

0549e1f3

[TensorPipe/RPC] Serialize and deserialize message (#36197)

jiayisuse committed 4 years ago

3411ec6e

[caffe2] L2 regularization for (RowWise)SparseAdagrad fusion on GPUs (#37805)

jianyuh committed 4 years ago

7fa897ea

[BE] Split pytorch_linux_test into 3 steps (#37808)

malfet committed 4 years ago

429d90f6

Add several ops for portal NLU/ASR model (again) (#37801)

linbinyu committed 4 years ago

458134f0

Make a separate cmake option for caffe2 tests (#37721)

suo committed 4 years ago

aff92ef3

add qnnpack path for hardtanh (#35779)

vkuzo committed 4 years ago

faad00a2

Remove std::complex to std::complex casting specialization (#37574)

zasdfgbnm committed 4 years ago

f5e6f39e

[Onnxifi] Cache output shape inference result for OnnxifiOp (#37796)

Yinghai Lu committed 4 years ago

15df33f7

Enable HgemmBatched for ROCm (#37483)

Huang Shuang committed 4 years ago

18455450

fix ROCm bench CI by increasing first iter timeout (#37633)

jeffdaily committed 4 years ago

4a2c642e

Math functions of c10::complex should be overloaded as const reference (#37689)

zasdfgbnm committed 4 years ago

090ea775

[FakeLowp] Reset workspace in test (#37799)

Yinghai Lu committed 4 years ago

8e5f162b

Use `gpu_kernel` in Affine Quantizer (#37312)

crcrpar committed 4 years ago

1d43d7ca

docs: Fixed docstring indentation for documentation (#37739)

frgfm committed 4 years ago

847d102e

Newer Older