strisunshinewentingwang/pytorch

Pull Requests Commits

Extend autograd functional benchmarking to run vectorized tasks (#67045)

soulitzer committed 2 years ago

21c6de9f

[shard] fix named_params_with_sharded_tensor (#70228)

wanchaol committed 2 years ago

82c5f298

[DataPipe] adding a finally statement to ensure hook is reset (#70214)

NivekT committed 2 years ago

74c834e0

Fixed typo in torch check for cdist (#70178)

vfdev-5 committed 2 years ago

23902fb8

Implementation of Wishart distribution (#68588)

nonconvexopt committed 2 years ago

a217a62e

[reland] Support torch.equal for ShardedTensor. (#70145)

pritamdamania committed 2 years ago

0544f975

[Operator Versioning] Split the upgrader test to a separate file and cover mobile part (#70090)

cccclai committed 2 years ago

c321d4c1

[StaticRuntime] Add TensorExpr fusion with dynamic shapes in SR (#69475)

navahgar committed 2 years ago

a6f95315

[jit] Add support for dynamic shape fusion in JIT. (#69474)

navahgar committed 2 years ago

c6d11623

[nnc] tensorexpr for quantized::add (#70188)

IvanKobzarev committed 2 years ago

c5333cdf

bug fix FractionalMaxPool2d (random_samples dimensions) (#70031)

george-qi committed 2 years ago

bb515199

[StaticRuntime] Refactor StaticModule to pass in sample inputs (#69473)

navahgar committed 2 years ago

91da2d5f

fix cpu binary size increase for clamp (#70168)

Natalia Gimelshein committed 2 years ago

c4a6c7a4

[nnc] Move DispatchParallel to external_functions (#70221)

IvanKobzarev committed 2 years ago

5504e4ae

Change TH_BLAS_MKL into AT_MKL_ENABLED() (#70219)

peterbell10 committed 2 years ago

304efd8e

[FSDP/Checkpoint] Activation offload support in checkpoint_wrapper (#70165)

rohan-varma committed 2 years ago

a197f3fe

Android build migrated to GHA. (#68843)

b0noI committed 2 years ago

e428a905

Revert "Revert D32498572: allow external backend codegen to be used without autograd kernels" (#69949)

bdhirsh committed 2 years ago

5e222d08

Add explicit OperatorHandle destructor (#70033)

esaliya committed 2 years ago

8e763cd7

dbr quant: better fix for bug with recursion on dequantize (#70128)

vkuzo committed 2 years ago

adaf3838

dbr quant: stop overridding tensor getters (#70115)

vkuzo committed 2 years ago

cce9c9aa

dbr quant: clean up logging format (#70114)

vkuzo committed 2 years ago

f2917080

dbr quant: add test for qconfig_dict and methods (#70109)

vkuzo committed 2 years ago

fb2a6747

update example in classification losses (#69816)

rohitgr7 committed 2 years ago

78bea1bb

Revert D33241684: [pytorch][PR] Install TensorRT lib on oss docker and enable fx2trt unit test

suo committed 2 years ago

19f89840

Remove backward op for slow dilated 2d convolution (#70067)

jbschlosser committed 2 years ago

b376d82c

Install TensorRT lib on oss docker and enable fx2trt unit test (#70203)

wushirong committed 2 years ago

dab3d313

[fusion] Add ConvTranspose+BN fusion support (#70022)

jonmorton committed 2 years ago

123be0e5

[Static Runtime] Support native op split_with_sizes (#69999)

donaldong committed 2 years ago

24f16de9

Handle the corner case when min == max in L2 search (#70207)

Haixin Liu committed 2 years ago

6623c483

Newer Older