strisunshinewentingwang/pytorch

Pull Requests Commits

[caffe2] exposes Net cancellation through pybind state (#44043)

dahsh committed 4 years ago

5ee31308

Fix HashStoreTests and move to Gtest (#43384)

osalpekar committed 4 years ago

e028ad07

Modularize FileStoreTest and move to Gtest (#43383)

osalpekar committed 4 years ago

69a3ff00

Convert StoreTestUtils to Gtest (#43382)

osalpekar committed 4 years ago

a7fba7de

Improving ModuleList indexing error msg (#43361)

Elias Ellison committed 4 years ago

b69c28d0

use non-overflowing divide in cuda kernel util GET_BLOCKS (#44391)

Rong Rong committed 4 years ago

c010ef7f

[pyper] export caffe2 bucketize GPU operator to pytorch

xw285cornell committed 4 years ago

ba6ddaf0

Revert D23568330: [pytorch][PR] Moves some of TestTorchMathOps to OpInfos

Elias Ellison committed 4 years ago

e0c65abd

Small fixes in Dependency.cmake and run_test.py (#44414)

malfet committed 4 years ago

fc51047a

[JIT] Support partially specified sizes/strides in IRParser (#44113)

Lilyjjo committed 4 years ago

b0bcdbb1

[quant] quantized path for ConstantPadNd (#43304)

z-a-f committed 4 years ago

36742649

fix typo in embedding_bag_non_contiguous_weight test (#44382)

lixinyu committed 4 years ago

032480d3

[PyTorch][Mobile] Insert the module name as `name()` to metadata dict if metadata doesn't contain "model_name" (#44400)

husthyc committed 4 years ago

a00d36b0

Check commutativity for computed dispatch table and add a test to check entries. (#44088)

Ailing Zhang committed 4 years ago

24efd29d

[NCCL] Add Environment Variable to guard Async Error Handling feature (#44163)

osalpekar committed 4 years ago

48c47db8

[NCCL] ProcessGroupNCCL Destructor Blocks on WorkNCCL Completion (#41054)

osalpekar committed 4 years ago

211ece72

[NCCL] WorkNCCL Helper Functions (#41053)

osalpekar committed 4 years ago

afbf2f14

[NCCL] Abort Errored and Timed Out NCCL Communicators from Watchdog Thread (#41052)

osalpekar committed 4 years ago

f8f7b784

[NCCL] Use cudaEventQuery to Poll for GPU operation errors (#41051)

osalpekar committed 4 years ago

4e5c55ef

[NCCL] Timeout Loop Thread for Async Error Handling (#41050)

osalpekar committed 4 years ago

1df24fd4

Preserve .ninja_log in build artifacts (#44390)

malfet committed 4 years ago

15cbd1cf

[Reland] Optimize code path for adaptive_avg_pool2d when output size is (1, 1) (#44211)

xwang233 committed 4 years ago

ef4475f9

Benchmarks: make fuser and executor configurable from command line. (#44291)

Mikhail Zolotukhin committed 4 years ago

37093f4d

Misc. FakeLowP OSS cleanup (#44331)

venkatacrc committed 4 years ago

364d03a6

BUG: make cholesky_solve_out do broadcast, error checking (#43137)

mattip committed 4 years ago

758c2b96

Use compile time cudnn version if linking with it statically (#44402)

malfet committed 4 years ago

683380fc

Fix frac in CUDA fuser (#44152)

bertmaher committed 4 years ago

6ec8fabc

Prevent the TE fuser from getting datatypes it can't handle (#44160)

bertmaher committed 4 years ago

350130a6

[te] Fix casting of unsigned char, and abs(int) (#44157)

bertmaher committed 4 years ago

960c088a

Skipping CUDA tests in ProcessGroupGloo and logs (#42488)

osalpekar committed 4 years ago

7c464eed

Newer Older