strisunshinewentingwang/pytorch

Pull Requests Commits

torch.complex and torch.polar (#39617)

wjgan7 committed 4 years ago

e4373083

[TensorExpr] Cleanup logic in the TensorExpr fuser pass. (#42938)

Mikhail Zolotukhin committed 4 years ago

b9a105bc

[TensorExpr] Remove redundant checks from canHandle in TE fuser. (#42937)

Mikhail Zolotukhin committed 4 years ago

fc304bec

[TensorExpr] Wrap fuser in a class. (#42936)

Mikhail Zolotukhin committed 4 years ago

48c183af

Reconstruct scopes (#41615)

taivu committed 4 years ago

02c8ad70

Add more verbose error message about PackedSequence lengths argument (#42891)

malfet committed 4 years ago

3dc84531

Clearer Semantics and Naming for Customized Quantization Range Initialization in Observer (#42602)

Paul Shao committed 4 years ago

b992a927

[reland][quant][fix] Remove activation_post_process in qat modules (#42343) (#43015)

jerryzh168 committed 4 years ago

a55b7e2a

Back out "change pt_defs.bzl to python file"

ljk53 committed 4 years ago

8cf01c5c

Python/C++ API Parity: TransformerDecoderLayer (#42717)

VinodSKumar committed 4 years ago

830423b8

[quant][doc] Print more info for fake quantize module (#43031)

jerryzh168 committed 4 years ago

85752b98

[jit][static runtime] Simplify the graph and add operator whitelist (#43024)

bwasti committed 4 years ago

523b2ce9

Allow RPC to be initialized again after shutdown. (#42723)

pritamdamania committed 4 years ago

89b0b3bc

Nightly checkout tool (#42635)

scopatz committed 4 years ago

21823aa6

Add DDP+RPC tutorial to RPC docs page. (#42828)

pritamdamania committed 4 years ago

a6b69fdd

make deadline=None for all numerics tests (#43014)

Hector Yuen committed 4 years ago

3544f60f

Fix to Learnable Fake Quantization Op Benchmarking (#43018)

Paul Shao committed 4 years ago

8b5642a7

Enable torch.utils typechecks (#42960)

malfet committed 4 years ago

6753157c

Add executor and fuser options to the fastrnn test fixture (#42946)

bertmaher committed 4 years ago

eb47940c

Update ort-nightly version to dev202008122 (#43019)

malfet committed 4 years ago

fd5ed4b6

[quant] Make PerChannel Observer work with float qparams (#42690)

supriyar committed 4 years ago

816d37b1

[quant] Create PerRowQuantizer for floating point scale and zero_point (#42612)

supriyar committed 4 years ago

6f844684

collect_env.py: Print CPU architecture after Linux OS name (#42961)

malfet committed 4 years ago

0ff51acc

Do not ignore `torch/__init__.pyi` (#42958)

malfet committed 4 years ago

ebc7ebc7

[NNC] Fix some bugs in Round+Mod simplification (#42934)

nickgg committed 4 years ago

6fb5ce55

update clone doc (#42931)

albanD committed 4 years ago

f03f9ad6

[tensorexpr] Autograd for testing (#42548)

bwasti committed 4 years ago

ba9025bc

Revert D22856816: [quant][fix] Remove activation_post_process in qat modules

zou3519 committed 4 years ago

607e49cc

Enroll TensorPipe agent in C++-only E2E test (#42680)

lw committed 4 years ago

8493b0d5

Remove Python dependency from TensorPipe RPC agent (#42678)

lw committed 4 years ago

c88d3a5e

Newer Older