dnnc-operators

Development and discussion of DNNC operators.

• Our work is based on ONNX operator release 1.5.0 till now.

• To get started on development, see this Developer's getting started guide.

• To get an idea of how to implement numpy like interface with dnnc see dnnc Tensor Functions.

• To see how to implement dnnc python interface see this Implementation guide

• To see why we are using Eigen library check this: Why should we use Eigen

• To refer back to the original Readme click here.

Below is the list of current ongoing operators list (Under Construction)

---

Operator	Doxygen	Testcases	Dimensions required	Dimensions supported	Broadcasting required	Broadcasting supported	Completion status	Contributer
Abs			N D	N D	❌	❌	✔️	Hrishikesh
Acos			N D	N D	❌	❌	✔️	Hrishikesh
Acosh			N D	N D	❌	❌	✔️	Hrishikesh
Add	✔️	✔️	N D	N D	✔️	✔️	✔️	Rohit
And	✔️	✔️	N D	N D	✔️	✔️	✔️	Gunjan
ArgMax
ArgMin
Asin			N D	N D	❌	❌	✔️	Hrishikesh
Asinh			N D	N D	❌	❌	✔️	Hrishikesh
Atan			N D	N D	❌	❌	✔️	Hrishikesh
Atanh			N D	N D	❌	❌	✔️	Hrishikesh
AveragePool
BatchNormalization
BitShift	✔️	✔️	N D	N D	✔️	✔️	✔️	Gunjan
Cast
Ceil
Clip
Compress
Concat
Constant			N D	N D	❌	❌	✔️	Rohit
ConstantOfShape
Conv		✔️	N D	N D			✔️	Rohit
ConvInteger
ConvTranspose
Cos
Cosh
CumSum
DepthToSpace
DequantizeLinear	✔️	✔️	N D	N D	❌	❌	✔️	Gunjan
Div	✔️	✔️	N D	N D	✔️	✔️	✔️	Gunjan
Dropout	✔️	✔️	N D	N D	❌	❌	✔️	Gunjan
Elu	✔️	✔️	1 D	1 D	❌	❌	✔️	Gunjan
Equal	✔️	✔️	N D	N D	✔️	✔️	✔️	Gunjan
Erf	✔️	✔️	N D	N D	❌	❌	✔️	Gunjan
Exp	✔️	✔️	N D	N D	❌	❌	✔️	Gunjan
Expand
EyeLike	✔️	✔️	2 D	2 D	❌	❌	DataType required	Gunjan
Flatten	✔️	✔️	N D	N D	❌	❌	✔️	Gunjan
Floor	✔️	✔️	N D	N D	❌	❌	✔️	Gunjan
GRU
Gather
GatherElements
Gemm	✔️	✔️	2 D	2 D	✔️	❌	✔️	Gunjan
GlobalAveragePool	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
GlobalLpPool	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
GlobalMaxPool	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
Greater	✔️	✔️	N D	N D	✔️	✔️	✔️	Nalin Shani
HardSigmoid	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
HardMax	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
Identity	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
If
InstanceNormalization	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
IsInf	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
IsNaN	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
LRN	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
LSTM
LeakyRelu	✔️	✔️	N D	N D	❌	❌	✔️	Nalin Shani
Less	✔️	✔️	N D	N D	✔️	✔️	✔️	Nikhil
Log	✔️	✔️	N D	N D	❌	❌	✔️	Nikhil
LogSoftmax	✔️	✔️	N D	N D	❌	❌	✔️	Nikhil
Loop
LpNormalization	✔️	✔️	2 D	2 D	❌	❌	✔️	Nikhil
LpPool
MatMul		4 D doesn't work	N D	4 D	❌	❌	✔️	Rohit
MatMulInteger	✔️	4 D doesn't work	N D	3 D	❌	❌	✔️	Nikhil
Max	✔️	✔️	N D	N D	✔️	✔️	✔️	Nikhil
MaxPool
MaxRoiPool
MaxUnpool
Mean	✔️	✔️	N D	N D	✔️	✔️	✔️	Nikhil
Min	✔️	✔️	N D	N D	✔️	✔️	✔️	Nikhil
Mod
Mul
Multinomial
Neg
NonMaxSupression
NonZero
Not	✔️	✔️	N D	N D	❌	❌	✔️	Gunjan
OneHot
Or	✔️	✔️	N D	N D	✔️	✔️	✔️	Gunjan
PRelu
Pad
Pow	✔️	✔️	N D	N D	✔️	✔️	✔️	Gunjan
QLinearConv
QlinearMatMul
QuantizeLinear
RNN
RandomNormal
RandomNormalLike
RandomUniform
RandomUniformLike
Reciprocal
ReduceL1
ReduceL2
ReduceLogSum
ReduceLogSumExp
ReduceMax
ReduceMean
ReduceMin
ReduceProd
ReduceSum
ReduceSumSquare
Relu
Reshape			N D	N D	❌	❌	✔️	Rohit
Resize
ReverseSequence
RoiAlign
Round
Scan
Scatter
ScatterElements
Selu
Shape
Shrink
Sigmoid			N D	3 D			✔️	Subham
Sign			N D	3 D			✔️	Subham
Sin			N D	3 D			✔️	Subham
Sinh			N D	3 D			✔️	Subham
Size
Slice
Softmax			2 D	N D			✔️	Subham
Softplus			N D	3 D			✔️	Subham
Softsign			N D	3 D			✔️	Subham
SpaceToDepth
Split
Sqrt			N D	3 D			✔️	Subham
Squeeze
StringNormalizer
Sub			N D	4 D	❌		✔️	Vishal
Sum
Tan			N D	3 D	❌		✔️	Vishal
Tanh			N D	3 D	❌		✔️	Vishal
TfIdfVectorizer
ThresholdedRelu			N D	N D	❌	❌	✔️	Rohit
Tile
TopK
Transpose			N D	3 D			✔️	Vishal
Unique
Unsqueeze
Upsample
Where
Xor	✔️	✔️	N D	N D	✔️	✔️	✔️	Gunjan