strisunshinewentingwang/pytorch

Pull Requests Commits

quant docs: add reduce_range explanatation to top level doc (#45305)

vkuzo committed 4 years ago

eb396243

Quantization: combine previous summary with new summary (#45135)

vkuzo committed 4 years ago

278da572

Quantization: add API summary section (#45093)

vkuzo committed 4 years ago

d2bd556e

[quant] conv_transpose graph patterns (#45078)

z-a-f committed 4 years ago

958c2086

Move xla codegen to aten. (#45241)

Ailing Zhang committed 4 years ago

606b1a9a

[dist_optim] introduce distributed functional optimizer (#45221)

wanchaol committed 4 years ago

32c355af

[optimizer] refactor Adam to use functional API (#44791)

wanchaol committed 4 years ago

08caf155

[optimizer] introduce optimizer functional API, refactor Adagrad (#44715)

wanchaol committed 4 years ago

0444c372

Enable `torch.cuda.nccl` typechecking (#45344)

malfet committed 4 years ago

8ab2ad30

Remove device maps from TensorPipe for v1.7 release (#45353)

mrshenli committed 4 years ago

5211fb97

adding a beta parameter to the smooth_l1 loss fn (#44433)

bdhirsh committed 4 years ago

439930c8

Use explicit templates in CUDALoops kernels (#44286)

malfet committed 4 years ago

37513a11

Add barrier() at the end of init_process_group and new_group. (#45181)

pritamdamania committed 4 years ago

a2b4177c

Add deprecation warning to PG backend and make TP backend stable. (#45356)

pritamdamania committed 4 years ago

3b7e4f89

[static runtime] Remove ops in static from backwards compatibility checks (#45354)

bwasti committed 4 years ago

04be4205

Add torch.do_assert, which is symbolically traceable (#45188)

vkuzo committed 4 years ago

eee7dad3

[RPC profiling] Add tests to ensure RPC profiling works on single threaded (#44923)

rohan-varma committed 4 years ago

7c5436d5

[RPC profiling] Extend RPC profiling to support async function execution over RPC. (#44664)

rohan-varma committed 4 years ago

27ab9bc0

Enable binary ops with Scalar Lists with for foreach APIs (#45298)

Iurii Zdebskyi committed 4 years ago

d5748d9a

Fix Windows build failure after DDP PR merged (#45335)

gunandrose4u committed 4 years ago

f07ac6a0

Add `torch.cuda.comm` to typechecking CI (#45350)

malfet committed 4 years ago

c8166d4b

port all JIT tests to gtest (#45264)

suo committed 4 years ago

22401b85

[pytorch] Update fmt to 7.0.3 (#45304)

suo committed 4 years ago

5a0514e3

CUDA BFloat16 neg (#45240)

zasdfgbnm committed 4 years ago

dc9e9c11

Add Deep and wide to test and flatten/tranpose for good measure (#44129)

bwasti committed 4 years ago

e5f6e5af

[static runtime] Add _out variants and reuse memory (#44128)

bwasti committed 4 years ago

d1a11618

[NNC] fix Half conversion of immediates in Cuda backend (#45213)

nickgg committed 4 years ago

d1d9017a

Vectorize bitwise_not (#45103)

xuhdev committed 4 years ago

536580e9

[quant][graph] Remove redundant aten::wait calls in the graph (#45257)

supriyar committed 4 years ago

a117d968

[ONNX] Correct a minor typo in warning (#45187)

shinh committed 4 years ago

8b00c4c7

Newer Older