This repository has been archived by the owner on Dec 26, 2024. It is now read-only.

Commits

History for

Commits on Sep 22, 2019

add async training support (#2 )
ymjiang
authored

Commits on Jun 25, 2019

apply BytePS patch
ymjiang
committed

Commits on Feb 27, 2019

[op] add back support for scalar type rescale_grad argument for adamw_update/mp_adamw_update (apache#14221 )
eric-haibin-lin
authored

Commits on Feb 26, 2019

[MXNET-1325] Make InferShapeAttr a standalone pass (apache#14193 )

junrushao
authored and
szha
committed

Commits on Feb 24, 2019

Dual stream cudnn Convolution backward() with MXNET_GPU_WORKER_NSTREAMS=2. (apache#14006 )

DickJC123
authored and
szha
committed

Commits on Feb 22, 2019

Add infer_type_partial (apache#14214 )

ptrendx
authored and
szha
committed

Commits on Feb 21, 2019

softmax for fp16 with fp32 accumulator (apache#14098 )

szha
authored and
eric-haibin-lin
committed

Commits on Feb 20, 2019

Commits on Feb 18, 2019

Refine runtime feature discovery python API and add documentation to … (apache#14130 )

larroy
authored and
szha
committed

Commits on Feb 16, 2019

Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN) (apache#13749 )

DickJC123
authored and
eric-haibin-lin
committed
Fix nd.pick large array issue (apache#14082 )

ChaiBapchya
authored and
eric-haibin-lin
committed

Commits on Feb 15, 2019

Parallelize CPU version and add GPU version of boolean_mask op (apache#14090 )

HyperZealot
authored and
zheng-da
committed
support dot(vector, vector) for fp16 inputs on GPU (apache#14102 )

eric-haibin-lin
authored and
szha
committed

Commits on Feb 14, 2019

Commits on Feb 13, 2019

add mkldnn softmax_output (apache#13699 )

rongzha1
authored and
TaoLv
committed

Commits on Feb 12, 2019

Addresses comments in runtime feature discovery API (apache#13964 )

larroy
authored and
szha
committed

Commits on Feb 11, 2019

Performance improvement in ToTensor GPU Kernel (apache#14099 )
sandeep-krishnamurthy
authored

Commits on Feb 10, 2019

[MKLDNN] Enable signed int8 support for convolution. (apache#13697 )

ZhennanQin
authored and
TaoLv
committed

Commits on Feb 8, 2019

Remove inplace support for ToTensor operator (apache#14083 )
sandeep-krishnamurthy
authored

Commits on Feb 6, 2019

Commits on Feb 5, 2019

CUDNN dropout (apache#13896 )

szha
authored and
eric-haibin-lin
committed

Commits on Feb 4, 2019

Image ToTensor operator - GPU support, 3D/4D inputs (apache#13837 )
sandeep-krishnamurthy
authored

Commits on Feb 1, 2019

Export resize and support batch size (apache#14014 )

stu1130
authored and
sandeep-krishnamurthy
committed

Commits on Jan 31, 2019

Fix inconsistent handling for FResourceRequestEx for imperative and symbolic executor (apache#14007 )
eric-haibin-lin
authored

Commits on Jan 30, 2019

Improve bulking in Gluon (apache#13890 )

ptrendx
authored and
eric-haibin-lin
committed

Commits on Jan 29, 2019

Image normalize operator - GPU support, 3D/4D inputs (apache#13802 )
sandeep-krishnamurthy
authored

Commits on Jan 28, 2019

fix compile error in debug mode (apache#13873 )

NEWPLAN
authored and
eric-haibin-lin
committed

Commits on Jan 27, 2019

MXNET-1295 Adding integer index support to Sequence* family of operators. (apache#13880 )

stephenrawls
authored and
szha
committed

Commits on Jan 26, 2019

[MXNET-703] Minor refactor of TensorRT code (apache#13311 )

KellenSunderland
authored and
sandeep-krishnamurthy
committed

Commits on Jan 25, 2019

api change (apache#13903 )

ZhennanQin
authored and
KellenSunderland
committed
Added optional parameters to BilinearResize2D to do relative scaling (apache#13985 )

ifeherva
authored and
sandeep-krishnamurthy
committed