strisunshinewentingwang/pytorch

Pull Requests Commits

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

Improve C++ maxpool and avgpool (#26521)

Will Feng committed 4 years ago

b5d15315

Typevar matching fix + implicit conversions from Scalar to int/float (#26453)

Basil Hosmer committed 4 years ago

167722d3

Quantized Interpolate Kernel(upsample_bilinear2d) (#26631)

lly-zero-one committed 4 years ago

03007b3d

Fix Future default constructor missing for ParallelNative

Bram Wasti committed 4 years ago

334e78b1

Export baddbmm (#25738)

neginraoof committed 4 years ago

63fd1054

Fix CI docker builds (#26704)

smessmer committed 4 years ago

3288da06

Validate Docker version in CI. (#26496)

ezyang committed 4 years ago

ae2a8fea

Prepare for Cocoapods 1.3 Release (#26751)

xta0 committed 4 years ago

f7742d2b

Remove one unnecessary copy of the output during the type promotion. (#26816)

VitalyFedyunin committed 4 years ago

f396b019

add function to get NCCL version for logging (#26583)

rohan-varma committed 4 years ago

d9055319

Newer Older