strisunshinewentingwang/pytorch

Pull Requests Commits

[ONNX] Squeeze operator should give an error when trying to apply to a dimension with shape > 1 (#38476)

yaeldMS committed 4 years ago

3c5e3966

Delete accidentally committed file errors.txt. (#43164)

gchanan committed 4 years ago

cd96dfd4

observers: use torch.all to check for valid min and max values (#43151)

vkuzo committed 4 years ago

57af1ec1

observers: use clamp instead of min/max in calculate_qparams (#43150)

vkuzo committed 4 years ago

3264ba06

observers: make eps a buffer (#43149)

vkuzo committed 4 years ago

a5dfba0a

quant bench: update observer configs (#42956)

vkuzo committed 4 years ago

5aa61afb

Remove unused variable vecVecStartIdx (#42257)

xuhdev committed 4 years ago

1f6e6a11

Use c10 threadpool for GPU to CPU distributed autograd continuations. (#42511)

pritamdamania committed 4 years ago

133e9f96

[jit] better error message (#43093)

ppwwyyxx committed 4 years ago

825ec18e

Fix type annotations for torch.sparse, enable in CI (#43108)

rgommers committed 4 years ago

864f0cfb

Adds movedim method, fixes movedim docs, fixes view doc links (#43122)

Mike Ruberry committed 4 years ago

6db0b878

Implement batching rules for some unary ops (#43059)

zou3519 committed 4 years ago

37252e8f

vmap: fixed to work with functools.partial (#43028)

zou3519 committed 4 years ago

768c2a8c

.circleci: Copy LLVM from pre-built image (#43038)

seemethere committed 4 years ago

9c3f5795

Rename XLAPreAutograd to AutogradXLA. (#43047)

Ailing Zhang committed 4 years ago

7cb8d68a

Set default ATen threading backend to native if USE_OPENMP is false (#43067)

malfet committed 4 years ago

034e6727

Add torch.dot for complex tensors (#42745)

anjali411 committed 4 years ago

aab66602

Fix freeze_module pass for sharedtype (#42457)

kimishpatel committed 4 years ago

472f2913

prepare to split transformer header file (#43069)

lixinyu committed 4 years ago

269fdb5b

add training mode to mobile::Module (#42880)

ann-ss committed 4 years ago

248b6a30

Adds arccosh alias for acosh and adds an alias consistency test (#43107)

Mike Ruberry committed 4 years ago

e2eb0cb1

Optimize SiLU (Swish) op in PyTorch (#42976)

xiaomengy committed 4 years ago

4ae832e1

Updates torch.clone documentation to be consistent with other functions (#43098)

Mike Ruberry committed 4 years ago

d4c5f561

Implement hstack, vstack, dstack (#42799)

muthuArivoli committed 4 years ago

5bcf9b01

[jit] DeepAndWide benchmark (#43096)

Hao Lu committed 4 years ago

88641488

[JIT] Represent profiled types as a node attribute (#43035)

Elias Ellison committed 4 years ago

91f3114f

Document unavailable reduction ops with NCCL backend (#42822)

rohan-varma committed 4 years ago

19902f6c

Add set_device_map to TensorPipeOptions to support GPU args (#42637)

mrshenli committed 4 years ago

06aaf8c2

Fix type annotations for a number of torch.utils submodules (#42711)

rgommers committed 4 years ago

c84f7847

Stop treating ASAN as special case (#43048)

malfet committed 4 years ago

bcf54f94

Newer Older