strisunshinewentingwang/pytorch

Pull Requests Commits

Remove gradient value as input from SparseNormalize op (#24357)

ffjiang committed 5 years ago

d7c6debc

For int64_t atomicAdd, use the available compiler builtin on ROCm. (#24854)

iotamudelta committed 5 years ago

9ebdf019

Allow SyncBatchNorm without DDP in inference mode (#24815)

ppwwyyxx committed 5 years ago

927fb56e

Fix VaryingShape::merge

zdevito committed 5 years ago

a04f729b

Add resnext 32x4d shapes to benchmark (#24503)

rohan-varma committed 5 years ago

60518e00

Change kernel_size to self.kernel_size to resolve error in quantized conv module (#24499)

rohan-varma committed 5 years ago

a6a13e36

Build libtorch binary with new ABI (#23908)

Will Feng committed 5 years ago

5aa0f89d

Use snake names for all files in distributed.rpc (#24502)

mrshenli committed 5 years ago

b6803d62

enable "keeps" from BoxWithNMSLimit and caffe2_fastrcnn_outputs_inference

wat3rBro committed 5 years ago

3b22bbeb

Cache node operators to speed up optimization (#24827)

suo committed 5 years ago

c6617b37

Update docs for softmax in onnx supported operators (#24832)

hackeritchy committed 5 years ago

c0a796d9

ShahriarSS committed 5 years ago

cd622f76

Add align_corners option to grid_sample and affine_grid, change default to False (#23923)

bnehoran committed 5 years ago

87217cfd

Remove unused files from THNN and THCUNN (#24820)

skrah committed 5 years ago

9e7083d0

Remove support for old architectures in cpp_extension and CMake (#24442)

rgommers committed 5 years ago

92c63d90

big cpp test reorg (#24801)

suo committed 5 years ago

dfdb86a5

Record function name as an attribute of CallFunction nodes.

Mikhail Zolotukhin committed 5 years ago

85564c14

Modify symmetric eigendecomposition derivative (#23018)

vishwakftw committed 5 years ago

9228dd76

Added .pyi file for flatten (#24459)

Chillee committed 5 years ago

5a032f02

Don't require slow test reporting in `run_tests.py --pytest` (#24448)

suo committed 5 years ago

0ce7264e

extra_repr for quantized modules (#24443)

James Reed committed 5 years ago

a0b13b4f

Use c10::ThreadPool to send and receive messages (#23968)

mrshenli committed 5 years ago

99dea08e

Enables `inplace` in the quantized relu (#24374)

z-a-f committed 5 years ago

dd97743d

Add `@ignore` for script classes (#23614)

davidriazati committed 5 years ago

aed306dc

Clear recursive error stack on each compilation (#23458)

davidriazati committed 5 years ago

10c45641

Enabled torch.mm and torch.mv for bfloat16

izdeby committed 5 years ago

eee3e929

Module: add dump function that recursively prints contents of the module. (#24356)

Mikhail Zolotukhin committed 5 years ago

cf57f73c

jit_log: Extract a function that prefixes all lines of a string with another string. (#24355)

Mikhail Zolotukhin committed 5 years ago

9b73c773

Respect pre-defined DOCKER_IMAGE value in binary_populate_env.sh (#24787)

Will Feng committed 5 years ago

76716f6c

fix double copying of constants (#24412)

Elias Ellison committed 5 years ago

2e44630d

Newer Older