strisunshinewentingwang/pytorch

Pull Requests Commits

Fix _empty_per_channel_affine_quantized to be less hacky (#26243)

Dmytro Dzhulgakov committed 5 years ago

9aad4d7b

adding OpProfile proto into ProfDAGProtos to support storing operation cost (#26677)

wenqicaofb committed 5 years ago

fbc3c148

Quantized Interpolate Kernel(upsample_nearest2d) (#26617)

lly-zero-one committed 5 years ago

ba8002ec

_per_channel_affine_qtensor -> _make_per_channel_quantized_tensor (#26679)

jerryzh168 committed 5 years ago

aa95c795

Skip observing bias across function call hierarchy (#26642)

jerryzh168 committed 5 years ago

8a919f4f

Whitelist ATen/core sources and headers for Caffe2 (#26609)

dreiss committed 5 years ago

af96e0cb

_per_tensor_affine_qtensor -> _make_per_tensor_quantized_tensor (#26678)

jerryzh168 committed 5 years ago

d63143dc

Add ObserveHelper and remove some common function parameters (#26641)

jerryzh168 committed 5 years ago

7d612066

fix github actions for forked PRs (#26562)

suo committed 5 years ago

7f89464b

Update qengine flag in python to string (#26620)

supriyar committed 5 years ago

45391cce

remove unneeded code (#26640)

jerryzh168 committed 5 years ago

5d82cefa

A few hub improvements (#25980)

Ailing Zhang committed 5 years ago

1eaaf8b6

Update ONNX Export for Gather and Scatter for Opset 11

lara-hdr committed 5 years ago

c79d116a

Fix Exporting RNN/LSTM's Initial State (h0/c0) to ONNX

lara-hdr committed 5 years ago

3569a1c6

quantized torch.topk (#26486)

James Reed committed 5 years ago

cb9fd0ce

Allow batch size of 0 in Conv

James Reed committed 5 years ago

64d58c2f

add CondValue to unify refinements and code emission (#26145)

zdevito committed 5 years ago

fcd13549

Fix the flaky test_qlinear test caused by hypothesis deadline (#26663)

jianyuh committed 5 years ago

cbdbdd3c

"fixing" gcc bug introduced with cuda 10.1 (#26445)

Mingbo Wan committed 5 years ago

21314cfd

Serialization for per channel qtensor (#26339)

Dmytro Dzhulgakov committed 5 years ago

ebc2365f

NHWC specialization for quantized::cat (#26524)

James Reed committed 5 years ago

c0aa6a01

nightly prefix for android nightly jobs (#26652)

IvanKobzarev committed 5 years ago

69631c3e

BlobReference __getattr__ can only throw AttributeError (#26654)

Aapo Kyrola committed 5 years ago

aeb6532e

Import torch.quantization when one imports torch

Dmytro Dzhulgakov committed 5 years ago

8fc86525

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

zou3519 committed 5 years ago

567a1981

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

ezyang committed 5 years ago

fdf2bdef

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

zou3519 committed 5 years ago

e0f86f7a

Makes test_indexing.py device generic (#26634)

Mike Ruberry committed 5 years ago

a9a9d362

Sync docker images

IvanKobzarev committed 5 years ago

2a574e49

Add testing script for iOS x86 build (#26632)

xta0 committed 5 years ago

781f8618

Newer Older