strisunshinewentingwang/pytorch

Pull Requests Commits

Enable batch_size = 0 support in DNNLOWP Concat operator (#26849)

jspark1105 committed 4 years ago

5231699d

remove backward functions from jit-op-registry for mobile build (#26851)

ljk53 committed 4 years ago

60372dc7

add mobile friendly at:parallel_for backend

ljk53 committed 4 years ago

ed260748

Improvements to GuardElimination and InsertBailouts

Krovatkin committed 4 years ago

14d7d571

Updating submodules

svcscm committed 4 years ago

20ed6ba0

Remove unnecessary functions and cleanup code in quantization.cpp.

Mikhail Zolotukhin committed 4 years ago

058ba0e7

Fix building with PARALLEL_BACKEND=NATIVE_TBB (#26742)

Arnaud Bonnet committed 4 years ago

8f359a48

Remove three unused declaration. (#26699)

xuhdev committed 4 years ago

c25c507f

Try to disable annoying hypothesis warnings again (#26853)

James Reed committed 4 years ago

f37aa2de

Re-write of tensor-scalar quantized add

James Reed committed 4 years ago

20ebd13f

Fix broken failure messages for OverloadedMethodValue

James Reed committed 4 years ago

1d55616a

Throw if someone tries to torch.save() quantized modules (#26828)

James Reed committed 4 years ago

df16fb9c

Remove convert_to_ssa argument from runCleanupPasses - it is only used in one place.

Mikhail Zolotukhin committed 4 years ago

d842435c

Use optimized_graph in graph_executor.

Mikhail Zolotukhin committed 4 years ago

9df887df

QEngine::QNNPACK enabled, module.eval()

IvanKobzarev committed 4 years ago

ed82a28c

Migrate multinomial from the TH to Aten (CUDA) (#26481)

ifedan committed 4 years ago

2eb59232

enable double backward for non-cudnn LSTM and GRU (#26660)

ngimel committed 4 years ago

90ffab6e

Move the CUDA implementation of log to ATen. (#26494)

xuhdev committed 4 years ago

91549ef6

Bytecode export flow (#25187)

Martin Yuan committed 4 years ago

7fc06ea5

Highlighting in the doc that square root comes before adding epsilon

vincentqb committed 4 years ago

660d9e24

Fix CUDA named tensor `copy_` (#26829)

zou3519 committed 4 years ago

08425d8c

support iterables, rangevalue in list comprehensions (#26768)

Elias Ellison committed 4 years ago

d43480d6

Do not call cpuinfo_initialize() on other than x86 arch. (#26265)

abalib committed 4 years ago

a23109e1

fix to operate on cuda kernel with clang and libc++ (#25553)

jcooky committed 4 years ago

93836015

torch.load default encoding change to 'utf-8' (#26421)

nmilosev committed 4 years ago

5fc52482

Named tensor support for: all, any, bitwise_not, cumprod, cumsum, and more (#26815)

zou3519 committed 4 years ago

92a2d423

Revert D17473200: [pytorch][distributed] add function to get NCCL version for logging

sampepose committed 4 years ago

4bd1da14

Convert TensorIterator to use function_ref, a lightweight alternative to std::function. (#26592)

Owen Anderson committed 4 years ago

81bbb7eb

Cuda101 upgrade (#26823)

Mingbo Wan committed 4 years ago

5379e87a

Revert D17565828: [pytorch][PR] [ONNX] Export baddbmm

houseroad committed 4 years ago

b6a1d618

Newer Older