strisunshinewentingwang/pytorch

Pull Requests Commits

[JIT] fix alias assertion (#36178)

Elias Ellison committed 4 years ago

8cb19508

[ONNX] disable size optimizations for onnx (#36243)

BowenBao committed 4 years ago

48bf3eef

Base class for the quantized ConvTranspose (#35370)

z-a-f committed 4 years ago

c5662dd5

[pt]Supported benchmarking pytorch jit self-contained models. (#35279)

newstzpz committed 4 years ago

7374a00b

[quant][fix] at::print for per channel affine quantized tensors (#36280)

jerryzh168 committed 4 years ago

f2bae8e8

Updating submodules

svcscm committed 4 years ago

51456dc8

[quant] Move graph mode quantization tests to test_quantize_script.py (#36324)

jerryzh168 committed 4 years ago

358466f1

Appropriately handle exceptions in autograd engine. (#36019)

pritamdamania committed 4 years ago

14ce500a

Fix `torch.min` docs (#36319)

zou3519 committed 4 years ago

9662ef66

Updating submodules

svcscm committed 4 years ago

1ffc2d9a

[RELAND] Add temporary impl_UNBOXED syntax sugar for unboxed-only defs. (#36223)

ezyang committed 4 years ago

2de3e491

[RELAND] Add DispatchKey impl overload; remove use of torch::dispatch (#36222)

ezyang committed 4 years ago

ef07bb65

[TensorExpr] add simplication of constant branches to IR Simplifier (#36257)

nickgg committed 4 years ago

477f1c04

caffe2/core/plan_executor: add cancellation of async nets on error + propagate exceptions via std::exception_ptr for stack traces (#31966)

d4l3k committed 4 years ago

90c7db8a

Revert D20768930: add quantized layer norm implementation

ezyang committed 4 years ago

88c22070

[quant][onnx] Mark upsample_nearest2d, sigmoid and reshape as no scale (#36325)

supriyar committed 4 years ago

d51ad40f

New CUDA Fuser code lowering refactor (#36199)

csarofeen committed 4 years ago

e551bfc8

Support for pruning delays in Adagrad Optimizer (#34527)

Dhruv Choudhary committed 4 years ago

f0ea6862

caffe2: preserve python exception type from PythonOp (#36267)

d4l3k committed 4 years ago

376542c8

remove some code part never been called (#35033)

huangzhiyuan committed 4 years ago

8493383e

Fixing RPC Shutdown and Thread Joining (#36239)

osalpekar committed 4 years ago

264da24c

Grad input padding support for dilation argument (#33872)

albanD committed 4 years ago

9497b21e

Revert D18672405: Revert D18672405: Use codegen'ed unboxing wrappers (#36010)

smessmer committed 4 years ago

e311e53a

[pt][quant] Add vector path to copy kernel for quantized data types (#36189)

dskhudia committed 4 years ago

866227cf

[TensorExpr] fix bug in IRSimplifier when multiplying by 0 (#36287)

nickgg committed 4 years ago

1443db8d

Revert "Revert D20885968: [pytorch][PR] Enable backtrace with MSVC" (#36205)

peterjc123 committed 4 years ago

5061ef63

make vendor match with this implementation (#36302)

Hector Yuen committed 4 years ago

423b0143

add quantized layer norm implementation (#35329)

vkuzo committed 4 years ago

f813e718

Add avx2 integer horizontal sum and sum of squares to vec256 qint types (#35693)

vkuzo committed 4 years ago

23e5f6a7

[pytorch] move force schema registration output into a separate file (#36284)

ljk53 committed 4 years ago

126d00c8

Newer Older