strisunshinewentingwang/pytorch

Pull Requests Commits

Import torch.quantization when one imports torch

Dmytro Dzhulgakov committed 4 years ago

8fc86525

Fix ellipsis behavior for `Tensor.align_to` to glob all missing dims (#26648)

zou3519 committed 4 years ago

567a1981

Revert D17450502: [pytorch][PR] [WIP] Enabled bfloat16 dtype on CUDA

ezyang committed 4 years ago

fdf2bdef

Add namedtensor build & tests to default sets (#26633)

zou3519 committed 4 years ago

e0f86f7a

Makes test_indexing.py device generic (#26634)

Mike Ruberry committed 4 years ago

a9a9d362

Sync docker images

IvanKobzarev committed 4 years ago

2a574e49

Add testing script for iOS x86 build (#26632)

xta0 committed 4 years ago

781f8618

fix operator level benchmark to have NHWC layout (#26577)

dskhudia committed 4 years ago

fc926d92

Simplify observers declaration with functools.partial (#26492)

Dmytro Dzhulgakov committed 4 years ago

a79b3685

Enabled bfloat16 dtype on CUDA

izdeby committed 4 years ago

76697a3b

prevent generating caffe2::mkldnn for multiple times (#25257)

jasjuang committed 4 years ago

e4821012

ATen port of lgamma (cuda) (#26600)

pearu committed 4 years ago

786d2259

Remove deprecated torch.gels (#26480)

vishwakftw committed 4 years ago

15b50606

Fixing the calling parameters of write_gif function of the moviepy.

danielmatte committed 4 years ago

557246b7

Revert D17521607: Name inference for min(Tensor, dim?) / max(Tensor, dim?)

zou3519 committed 4 years ago

808f4a4d

Renames `tensor.renamed -> rename`, `tensor.names_ -> rename_` (#26548)

zou3519 committed 4 years ago

4fada962

Name inference for min(Tensor, dim?) / max(Tensor, dim?) (#25582)

zou3519 committed 4 years ago

d3e90bc4

C++ API parity: at::Tensor::detach

pbelevich committed 4 years ago

6b255624

Whenever possible, use function pointers rather than std::function to represent Operation's. (#26560)

Owen Anderson committed 4 years ago

bdf10380

Unify Quantization APIs for add, pool and relu (#26586)

supriyar committed 4 years ago

99226cd5

Move unpickler related codes from pickler.h/cpp to unpickler.h/cpp (#26432)

Martin Yuan committed 4 years ago

7e619650

_dequantize_linear -> _dequantize_per_tensor (#26576)

jerryzh168 committed 4 years ago

95cb22f2

quantized average_pool2d and adaptive_avg_pool2d implementation(Revert d17437015) (#26580)

lly-zero-one committed 4 years ago

eca01eb0

Skip some fragile tests (#26599)

smessmer committed 4 years ago

fcfca9ad

quantize_linear_per_channel -> quantize_per_channel (#26575)

jerryzh168 committed 4 years ago

2e82ee03

Expose supportedQEngines to python (#26474)

jerryzh168 committed 4 years ago

2667493f

C++ API parity: at::Tensor::version

pbelevich committed 5 years ago

d117842e

Fix CI (#26593)

smessmer committed 5 years ago

aa785234

Add inplace argument to InsertObservers and InsertQuantDeQuant (#26389)

jerryzh168 committed 5 years ago

d09d1d9a

Get scalar type from observer module (#26425)

jerryzh168 committed 5 years ago

1bec8d7a

Newer Older