Versions Compared

Old Version 11

changes.mady.by.user Anirudh Acharya

Saved on

compared with

New Version Current

changes.mady.by.user Vandana Kannan

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • import of ONNX models into Mxnet.
  • export MXNet models to ONNX format.


ONNX-MXNET Import Module

We have two repositories for the import module:

  1. onnx-mxnet repository inside onnx https://github.com/onnx/onnx-mxnet  (deprecated)
  2. A newly created and refactored module inside Mxnet -> contrib. https://github.com/apache/incubator-mxnet/tree/master/python/mxnet/contrib/onnx/onnx2mx


MXNET-ONNX Export module

A new module has been added to MXNet repository to export MXNet models to ONNX model format. https://github.com/apache/incubator-mxnet/tree/master/python/mxnet/contrib/onnx/mx2onnx

 

Operator Coverage:

This table keeps track of the status of all ONNX operators supported by Mxnet.
ONNX backend test script reports the coverage on the operators and attributes.


Status meaning:

OK = currently we support the operator

OK, in review = PR is out

NOT OK, under development = Operator is missing on MXNet backend or direct 1:1 mapping doesn’t exist

NOT OK = Not supported right now


        
OperatorTest CoverageMxnet (Import)Mxnet (Export)
AbsYesOKOK
AcosYesOKOK
AddYesOKOK
AndYes 

OK

OK
ArgMaxYesOKOK
ArgMinYes OKOK
Asin

Yes

OKOK
AtanYesOKOk
AveragePoolYesOKOK
BatchNormalizationYesOKOK
CastYesOKOK
CeilYes OKOK
ClipYesOKOK
ConcatYesOKOK
ConstantYesOK
ConvYesOKOK
ConvTransposeYes OKOK
CosYesOKOK
Crop

OK
DepthToSpaceYesOKNOT OK, under development
DivYesOKOK
Dropout  YesOKOK
EluYesOKOK
EqualYesOKOK
ExpYesOKOK
FlattenYesPARTIAL, only supports default axis=1OK
Floor YesOKOK
GRU
NOT OK , under development
GatherYesOK
GemmYesOKOK
GlobalAveragePoolYesOKOK
GlobalLpPool Yes OKOK, In review
GlobalMaxPoolYesOKOK
Greater YesOKOK
HardSigmoid YesOKOK
Hardmax YesOKNOT OK
Identity
NOT OKOK, In review
If


InstanceNormalization YesOK, In reviewOK
LRNYes OKOK
LSTM
NOT OK, under development
LeakyReluYesOKOK
LessYes OKOK
LogYesOKOK
LogSoftMaxYes OKOK
Loop


LpNormalization  Yes
OK, In review
LpPoolYesOK
LpPoolMatMulYesOKOK, In review
MaxYesOKOK
MaxPoolYesOKOK
MaxRoiPoolYes OKOK, In review
MeanYes OKOK
MinYesOKOK
MulYesOKOK
MultinomialYesOKOK
NegYesOKOK
NotYes OKOK
OrYes OKOK
PReluYesOKOK
PadYesOKOK
PowYesPartial, only supports default axis=1OK
RNN 
NOT OK, under development
RandomNormalYes OKOK
RandomNormalLikeYes OKOK
RandomUniformYes OKOK
RandomUniformLikeYes OKOK
ReciprocalYes OKOK
ReduceL1Yes OKOK
ReduceL2 YesOKOK
ReduceLogSum YesOKNOT OK
ReduceLogSumExp YesOKNOT OK
ReduceMaxYesOKOK
ReduceMeanYesOKOK
ReduceMinYesOKOK
ReduceProdYesOKOK
ReduceSumYesOKOK
ReduceSumSquareYes OKNOT OK
ReluYesOKOK
ReshapeYesOKOK
SeluYesOKOK
Shape
NOT OK, under developmentOK
SigmoidYesOKOK
SinYesOKOK
SizeYesOKOK
SliceYesPartial, supports axis=1OK
SoftmaxYesOKOK
SoftplusYesOKOK
Softsign YesOKOK
SpaceToDepth YesOKNOT OK, under development
SplitYesOKOK
SqrtYes OKOK
SqueezeYesOKOK
SubYesOKOK
SumYesOKOK
TanYesOKOK
TanhYesOKOK
Tile YesNOT OK, under developmentNOT OK, under development
TransposeTopKYesOK, in reviewOK, in review
TransposeXorYes OKOK
experimental ATenUnsqueeze YesNOT OK, under developmentexperimental Affine NOT OK, under developmentexperimental ConstantFill NOT OK, under developmentexperimental Crop NOT OK, under developmentIn reviewOK
UpsampleYesOK, in reviewOK, in review
XorYes OKOK








experimental ATen
NOT OK
experimental Affine
NOT OK
experimental ConstantFill
NOT OK
experimental GRUUnit
NOT OK
experimental GivenTensorFill
NOT OKexperimental FC NOT OK, under developmentexperimental GRUUnit NOT OK, under developmentexperimental GivenTensorFill NOT OK, under developmentexperimental Identity NOT OK, under development
experimental ImageScaler YesOK, In review
experimental MeanVarianceNormalizationLoopIndex
NOT OK
experimental MeanVarianceNormalization
NOT OK, under development
experimental ParametricSoftplus 
NOT OK, under development
experimental Scale
NOT OK, under development
experimental ScaledTanh 
NOT OK, under development
experimental ThresholdedRelu
NOT OK, under developmentexperimental Upsample  OK, In review

...



ONNX models coverage:

ONNX backend tests include below models. The table lists current status on Mxnet:

 


ONNX ModelsMxnet Support
bvlc_alexnetOK
densenet121OK
inception_v1OK, In review
inception_v2OK, In review
resnet50OK
shufflenetOK, In review
squeezenetOK
vgg16OK
vgg19OK
bvlc_googlenetOK
bvlc_caffenetOK
bvlc-rcnn-ilsvrc13OK

 

 

 

 

 

 

 

 

 

 

...