Overview ¶

This document lists the layers (or operators) which can be imported and converted. Supported operators allow to address a large range of classical topologies targeting a Mobile or IoT resource-constrained runtime environment: SqueezeNet, MobileNet V1 or V2, Inception, SSD MobileNet v1,…

Purpose of this document is to list the operators and their associated constraints or limitations, please refer to the original documentation for details on a given layer.

ONNX is an open format built to represent machine learning models. A part of a subset of operators from Opset 7, 8, 9, 10 up to 20 of ONNX 1.15 is supported.

This file was automatically generated.

ST Edge AI Core version : 4.0
137 operators found

Summary table ¶

Following table contains the list of the operators that can be imported, if the constraints or limitations are met.

supported optional fused activation (or non-linearity): gelu, linear, relu, quantized_relu, relu_n1_to_1, relu_0_to_1, leaky_relu, relu6, elu, selu, sigmoid, hard_sigmoid, hard_swish, exponential, tanh, softmax, softplus, softsign, abs, acos, acosh, asin, asinh, atan, atanh, ceil, clip, cos, cosh, erf, flexerf, exp, floor, identity, log, logistic, neg, logical_not, prelu, probit, reciprocal, relu_generic, relu_thresholded, round, sign, sin, sinh, softmax_zero, sqrt, swish, tan

operator	data types	constraints/limitations
Abs	float32	common
Acos	float32	common
Acosh	float32	common
Add	float32	common
And	bool	common
ArgMax	float32, int32	common
ArgMin	float32, int32	common
ArrayFeatureExtractor	float32	common, specific
Asin	float32	common
Asinh	float32	common
Atan	float32	common
Atanh	float32	common
AveragePool	float32	common, specific
BatchNormalization	float32	common, specific
BitShift	float32	common
Cast	bool, int8, uint8, float32	common, specific
Ceil	float32	common
Clip	float32	common
Concat	float32	common, specific
Constant	float32	common
ConstantOfShape	float32	common
Conv	float32	common, specific
ConvTranspose	float32	common, specific
Cos	float32	common
Cosh	float32	common
DepthToSpace	float32, int8, uint8	common, specific
DequantizeLinear	float32, int8, uint8	common, specific
Div	float32	common
Dropout	float32	common, specific
Einsum	float32	common, specific
Elu	float32	common
Equal	float32, bool	common
Erf	float32	common
Exp	float32	common
Expand	float32, int8, uint8	common, specific
Flatten	float32	common, specific
Floor	float32	common
Gather	float32	common, specific
GatherElements	float32	common, specific
GatherND	float32	common, specific
Gelu	float32	common
Gemm	float32	common
GlobalAveragePool	float32	common
GlobalMaxPool	float32	common
Greater	float32, bool	common
GreaterOrEqual	float32, bool	common
GRU	float32	common, specific
Hardmax	float32	common, specific
HardSigmoid	float32	common, specific
HardSwish	float32	common
Identity	float32	common
InPlaceAccumulatorV2	float32, bool	common, specific
InstanceNormalization	float32	common
LabelEncoder	float32	common, specific
LayerNormalization	float32	common
LeakyRelu	float32	common
Less	float32, bool	common
LessOrEqual	float32, bool	common
LinearClassifier	float32	common, specific
LinearRegressor	float32	common
Log	float32	common
LogSoftmax	float32	common, specific
LpNormalization	float32	common, specific
LRN	float32	common
LSTM	float32	common, specific
MatMul	float32	common, specific
Max	float32	common
MaxPool	float32	common, specific
Mean	float32	common
Min	float32	common
Mod	float32	common
Mul	float32	common
Neg	float32	common
Normalizer	float32	common, specific
Not	float32	common
Or	bool	common
Pad	float32	common, specific
Pow	float32	common
PRelu	float32	common, specific
QLinearAdd	uint8, int8	common
QLinearAveragePool	uint8, int8	common, specific
QLinearConcat	uint8, int8	common, specific
QLinearConv	int8, uint8	common, specific
QLinearGlobalAveragePool	uint8, int8	common
QLinearMatMul	int8, uint8	common
QLinearMul	uint8, int8	common
QuantizeLinear	float32, int8, uint8	common, specific
Range	float32	common
Reciprocal	float32	common
ReduceL1	float32	common
ReduceL2	float32	common
ReduceLogSumExp	float32	common
ReduceMax	float32	common
ReduceMean	float32	common
ReduceMin	float32	common
ReduceProd	float32	common
ReduceSum	float32	common
ReduceSumSquare	float32	common
Relu	float32	common
Reshape	float32	common, specific
Resize	float32	common, specific
Round	float32	common
Scale	float32	common
Scaler	float32, float	common, specific
ScatterND	float32	common, specific
Selu	float32	common
Shape	float32, int8, uint8, int32	common, specific
Sigmoid	float32	common
Sign	float32	common
Sin	float32	common
Sinh	float32	common
Size	float32, int8, uint8, int32	common, specific
Slice	float32, int8, uint8	common
Softmax	float32	common, specific
Softplus	float32	common
Softsign	float32	common
SpaceToDepth	float32, int8, uint8	common
Split	float32, int8, uint8	common, specific
Sqrt	float32	common
Squeeze	float32	common
Sub	float32	common
Sum	float32	common
SVMClassifier	float32	common, specific
SVMRegressor	float32	common, specific
Tan	float32	common
Tanh	float32	common
ThresholdedRelu	float32	common
Tile	float32, int8, uint8	common, specific
TopK	float32, int8, uint8	common
Transpose	float32, int8, uint8	common, specific
TreeEnsembleClassifier	float32	common, specific
TreeEnsembleRegressor	float32	common, specific
Unsqueeze	float32	common
Upsample	float32	common, specific
Where	float32, int8, uint8, int16, uint16, int32, uint32, bool	common
Xor	bool	common
ZipMap	float32	common, specific

Common constraints ¶

input and output tensors must be not dynamic.
- variable-length batch dimension (i.e. (None,)) is considered as equal to 1
- must not be greater than 6D
- dimension must be in the range [0, 65536[
- batch dimension is partially supported in the axis/axes parameters
data type for the weights/activations tensors must be:
- float32, int8, uint8
- only int32 for the bias tensor is considered
- for some operators, bool and binary types are also supported
1D operator is mapped on the respective 2D operator by adding a singleton dimension on the input: (12,3) -> (12, 1, 3)

Operators ¶

Abs ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Acos ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Acosh ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Add ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

And ¶

Performs boolean element-wise operation

category: eltwise operator
input data types: bool
output data types: bool

ArgMax ¶

Computes the indices of the max elements of the input tensor’s element along the provided axis.

category: generic layer
input data types: float32
output data types: int32

ArgMin ¶

Computes the indices of the min elements of the input tensor’s element along the provided axis.

category: generic layer
input data types: float32
output data types: int32

ArrayFeatureExtractor ¶

Select elements of the input tensor besed of the indices passed

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

used to support for example the skit-learn algo:
- Nu-Support Vector Classification (sklearn.svm.NuSVC)
- C-Support Vector Classification. (sklearn.svm.SVC)
- A random forest classifier (sklearn.ensemble.RandomForestClassifier)
- …

Asin ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Asinh ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Atan ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Atanh ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

AveragePool ¶

Downsamples the input

category: pooling layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

arbitrary strides, provided that they are smaller than the input size
arbitrary pool sizes, provided that they are smaller than the input size
The attribute dilations different from the default value is not supported

BatchNormalization ¶

Performs the normalization of the input

category: normalization layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

Only one output (Y) is supported

BitShift ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

Cast ¶

Cast elements of the input tensor to the specified output tensor data

category: conversion layer
input data types: bool, int8, uint8, float32
output data types: bool, int8, uint8, float32

Specific constraints/recommendations:

The attribute saturate different from the default value is not supported

Ceil ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Clip ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Concat ¶

Performs concatenation of a list of inputs

category: merge operator
input data types: float32
output data types: float32

Specific constraints/recommendations:

concatenating on the batch dimension is not supported

Constant ¶

input data types: float32
output data types: float32

ConstantOfShape ¶

Generates a tensor with given value and shape

category: constant layer
input data types: float32
output data types: float32

Conv ¶

Performs convolution operation

category: convolutional layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

arbitrary strides, provided that they are smaller than the input size
arbitrary filter kernel sizes, provided that they are smaller than the input size

ConvTranspose ¶

Transposed convolutional layer

category: convolutional layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

arbitrary strides, provided that they are smaller than the input size
arbitrary filter kernel sizes, provided that they are smaller than the input size

Cos ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Cosh ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

DepthToSpace ¶

Permutes the dimensions of the input according to a given pattern

category: reshaping layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Specific constraints/recommendations:

transposing the batch dimension is not supported

DequantizeLinear ¶

Computes element-wise data conversion low precision to full precision, based on the scale/zeropoint parameters

category: conversion layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Specific constraints/recommendations:

The attribute block_size different from the default value is not supported

Div ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

Dropout ¶

Applies Dropout to the input

category: regularization layers
input data types: float32
output data types: float32

Specific constraints/recommendations:

operator is dropped during the conversion

Einsum ¶

Einsum (Einstein Summation) operator

category: einsum operator
input data types: float32
output data types: float32

Specific constraints/recommendations:

diagonal is not supported

Elu ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Equal ¶

Performs logical element-wise operation

category: eltwise operator
input data types: float32
output data types: bool

Erf ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Exp ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Expand ¶

Constructs a tensor by tiling the input tensor

category: reshaping layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Specific constraints/recommendations:

tiling on batch-dimension is not supported

Flatten ¶

Flattens the non-batch input dimensions to a vector

category: Reshaping operation
input data types: float32
output data types: float32

Specific constraints/recommendations:

Flatten on the batch dimension is not supported
operator is dropped during the conversion

Floor ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Gather ¶

Gathers values along a specified axis

category: activation function
input data types: float32
output data types: float32

Specific constraints/recommendations:

Gather is not supported with indices dimensions > 2 (Batch is not considered), axis > 3and axis = 0, batch_dims attribute is not handled

GatherElements ¶

Gathers Elements along a specified axis

category: activation function
input data types: float32
output data types: float32

Specific constraints/recommendations:

GatherElements is not supported with indices dimensions > 2 (Batch is not considered), axis > 3and axis = 0, batch_dims attribute is not handled

GatherND ¶

Gathers slices from input tensor into an output tensor with shape specified by indices

category: activation function
input data types: float32
output data types: float32

Specific constraints/recommendations:

Batch_dims attribute is not handled. 4D or more indices (BATCH is not included) are not implementedLast dimension of indices > 4 with 2D inputs tensor case is not handled

Gelu ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Gemm ¶

General Matrix multiplication: Compute Y = alpha A B + beta*C

category: core layer
input data types: float32
output data types: float32

GlobalAveragePool ¶

Downsamples the input

category: pooling layer
input data types: float32
output data types: float32

GlobalMaxPool ¶

Downsamples the input

category: pooling layer
input data types: float32
output data types: float32

Greater ¶

Performs logical element-wise operation

category: eltwise operator
input data types: float32
output data types: bool

GreaterOrEqual ¶

Performs logical element-wise operation

category: eltwise operator
input data types: float32
output data types: bool

GRU ¶

Gated Recurrent Unit

category: recurrent layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

stateless and stateful (batch=1 only) mode support
fused activation: gelu, linear, relu, quantized_relu, relu_n1_to_1, relu_0_to_1, leaky_relu, relu6, elu, selu, sigmoid, hard_sigmoid, hard_swish, exponential, tanh, softmax, softplus, softsign
fused recurrent activation: gelu, linear, relu, quantized_relu, relu_n1_to_1, relu_0_to_1, leaky_relu, relu6, elu, selu, sigmoid, hard_sigmoid, hard_swish, exponential, tanh, softmax, softplus, softsign
return_state not supported

Hardmax ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

The value 1 is supported as the default value of the axis attribute, not the value -1

HardSigmoid ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

It is supported only for 1D tensor and only on the channel dimension

HardSwish ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Identity ¶

This layer should be used as a placeholder when no operation is to be performed.The layer just returns its inputs argument as output.

category: Identity layer
input data types: float32
output data types: float32

InPlaceAccumulatorV2 ¶

Performs an in-place accumulation (or overwrite) of a tensor into an accumulator buffer. This operator is typically used to gather/accumulate gradients or partial results in-place and exposes a control flag to select overwrite vs accumulate behavior.

category: utility / inplace operator
input data types: float32, float32, bool
output data types: bool, float32

Specific constraints/recommendations:

Input 0: accumulator buffer (same shape as Input 1).
Input 1: tensor to accumulate into the buffer (same dtype/shape as Input 0).
Input 2: scalar boolean overwrite flag; when true the buffer is overwritten with Input 1, otherwise Input 1 is added element-wise to the buffer.
Output 0: boolean success/status (single-element bool).
Output 1: updated accumulator buffer (same shape as Input 0).
All numeric tensors are expected to be float32 for the C implementation provided; integer/quantized formats are not supported.
Shapes of Input 0 and Input 1 must match. Input 2 is expected to be a scalar boolean.
This is an ONNX operator and not present in standard Keras/TensorFlow.

InstanceNormalization ¶

Apply instance normalization

category: normalization function
input data types: float32
output data types: float32

LabelEncoder ¶

Maps each element in the input tensor to another value.

category: LabelEncoder
input data types: float32
output data types: float32

Specific constraints/recommendations:

Warning: This operator is not supported for C code generation, it is only used during iForest generation. Support is not complete.
Attributes keys_floats, keys_strings, values_int64s, values_strings are not supported

LayerNormalization ¶

Apply LayerNormalization along the provided axis

category: normalization function
input data types: float32
output data types: float32

LeakyRelu ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Less ¶

Performs logical element-wise operation

category: eltwise operator
input data types: float32
output data types: bool

LessOrEqual ¶

Performs logical element-wise operation

category: eltwise operator
input data types: float32
output data types: bool

LinearClassifier ¶

LinearClassifier layer

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

used to support for example the skit-learn algo:
- Linear Support Vector Classification. (sklearn.svm.LinearSVC)
- Logistic Regression (aka logit, MaxEnt) classifier. (sklearn.linear_model.LogisticRegression)
- Classifier using Ridge regression. (sklearn.linear_model.RidgeClassifier)
- Linear Discriminant Analysis. (sklearn.discriminant_analysis.LinearDiscriminantAnalysis)

LinearRegressor ¶

General Matrix multiplication: Compute Y = alpha A B + beta*C

category: core layer
input data types: float32
output data types: float32

Log ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

LogSoftmax ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

It is supported only for 1D tensor and only on the channel dimension
The value 1 is supported as the default value of the axis attribute, not the value -1

LpNormalization ¶

Apply Lp-normalization along the provided axis

category: normalization function
input data types: float32
output data types: float32

Specific constraints/recommendations:

order of the normalization (1 and 2 supported)

LRN ¶

Apply Local Response Normalization over local input regions

category: normalization function
input data types: float32
output data types: float32

LSTM ¶

Computes a multi-layer long short-term memory (LSTM) RNN to an input sequence (batch=1, timesteps, features)

category: recurrent layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

stateless mode support only
fused activation: sigmoid
fused recurrent activation: sigmoid
return_state not supported
Only 1 input is allowed for LSTM; the others should be constant and placed into the initializers
layout=1 attribute is not supported

MatMul ¶

General Matrix multiplication: Compute Y = alpha A B + beta*C

category: core layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

Only up to 3D matrix multiplication is supported

Max ¶

Computes the maximum (element-wise) a list of inputs

category: eltwise operator
input data types: float32
output data types: float32

MaxPool ¶

Downsamples the input

category: pooling layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

arbitrary strides, provided that they are smaller than the input size
arbitrary pool sizes, provided that they are smaller than the input size
The attribute dilations different from the default value is not supported
The attribute storage_order different from the default value is not supported

Mean ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

Min ¶

Computes the minimum (element-wise) a list of inputs

category: eltwise operator
input data types: float32
output data types: float32

Mod ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

Mul ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

Neg ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Normalizer ¶

Normalize the input.

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

used to support for example the skit-learn algo:
- Logistic Regression (aka logit, MaxEnt) classifier. (sklearn.linear_model.LogisticRegression)
- Normalize samples individually to unit norm. (sklearn.preprocessing.Normalizer)
Note that the implementation of the max norm is aligned with sklearn implementation, not the onnx implementation. The difference is that sklearn is using absolute values to get a positive divisor when computing the norm. Onnx implementation does not use absolute values.

Not ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Or ¶

Performs boolean element-wise operation

category: eltwise operator
input data types: bool
output data types: bool

Pad ¶

Pads an input tensor

category: Reshaping layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

Axes are not supported for Pad operator
wrap value for mode attribute is not supported

Pow ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

PRelu ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

Only 1 input is allowed for PRelu; the slope should be constant and placed into the initializers

QLinearAdd ¶

Performs element-wise binary addition on 8 bit data types(with Numpy-style broadcasting support).

category: eltwise operator
input data types: uint8, int8
output data types: uint8, int8

QLinearAveragePool ¶

Downsamples an input tensor X and applies average pooling across the tensor according to kernel sizes, stride sizes, and pad lengths

category: pooling layer
input data types: uint8, int8
output data types: uint8, int8

Specific constraints/recommendations:

arbitrary strides, provided that they are smaller than the input size
arbitrary pool sizes, provided that they are smaller than the input size

QLinearConcat ¶

Concatenate a list of tensors into a single tensor

category: merge operator
input data types: uint8, int8
output data types: uint8, int8

Specific constraints/recommendations:

concatenating on the batch dimension is not supported

QLinearConv ¶

Consumes a quantized input tensor

category: convolutional layer
input data types: int8, uint8
output data types: int8, uint8

Specific constraints/recommendations:

arbitrary strides, provided that they are smaller than the input size
arbitrary filter kernel sizes, provided that they are smaller than the input size

QLinearGlobalAveragePool ¶

Downsamples an input tensor X and applies Average pooling acrossthe values in the same channel

category: pooling layer
input data types: uint8, int8
output data types: uint8, int8

QLinearMatMul ¶

Matrix product consuming two quantized input tensors

category: core layer
input data types: int8, uint8
output data types: int8, uint8

QLinearMul ¶

Performs element-wise binary multiplication on 8 bit data types(with Numpy-style broadcasting support).

category: eltwise operator
input data types: uint8, int8
output data types: uint8, int8

QuantizeLinear ¶

Computes element-wise data conversion full precision to low precision, based on the scale/zeropoint parameters

category: conversion layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Specific constraints/recommendations:

The attribute block_size different from the default value is not supported
The attribute saturate different from the default value is not supported
The attribute output_dtype different from the default value is not supported

Range ¶

Generate a tensor containing a sequence of numbers that begin at start and extends by increments of delta up to limit (exclusive).

category: generic layer
input data types: float32
output data types: float32

Reciprocal ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

ReduceL1 ¶

category: reduction operation
input data types: float32
output data types: float32

ReduceL2 ¶

category: reduction operation
input data types: float32
output data types: float32

ReduceLogSumExp ¶

category: reduction operation
input data types: float32
output data types: float32

ReduceMax ¶

Computes the Max of the input tensor’s element along the provided axes

category: reduction operation
input data types: float32
output data types: float32

ReduceMean ¶

Computes the Mean of the input tensor’s element along the provided axes

category: reduction operation
input data types: float32
output data types: float32

ReduceMin ¶

Computes the Min of the input tensor’s element along the provided axes

category: reduction operation
input data types: float32
output data types: float32

ReduceProd ¶

Computes the Product of the input tensor’s element along the provided axes

category: reduction operation
input data types: float32
output data types: float32

ReduceSum ¶

Computes the Sum of the input tensor’s element along the provided axes

category: reduction operation
input data types: float32
output data types: float32

ReduceSumSquare ¶

Computes the Sum Square of the input tensor’s element along the provided axes

category: reduction operation
input data types: float32
output data types: float32

Relu ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Reshape ¶

Reshapes a tensor

category: Reshaping operation
input data types: float32
output data types: float32

Specific constraints/recommendations:

The attribute allowzero different from the default value is not supported

Resize ¶

Resize the input tensor: it calculates every value in the output tensor as a weighted average of neighborhood

category: resizing operation
input data types: float32
output data types: float32

Specific constraints/recommendations:

Cubic resize is not supported
half_pixel_symmetric value for coordinate_tansformation_mode is not supported
The attribute antialias different from the default value is not supported
axes attribute is not supported
The attribute keep_aspect_ratio_policy different from the default value is not supported

Round ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Scale ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

Scaler ¶

Compute the Scale and Offset of input tensor

category: onnx.ml
input data types: float32
output data types: float

Specific constraints/recommendations:

used to support for example the skit-learn algo:
- Gradient Boosting for classification. (sklearn.ensemble.GradientBoostingClassifier)
- Transform features by scaling each feature to a given range. (sklearn.preprocessing.MinMaxScaler)
- Standardize features by removing the mean and scaling to unit variance. (sklearn.preprocessing.StandardScaler)
- Scale features using statistics that are robust to outliers. (sklearn.preprocessing.RobustScaler)
- Scale each feature by its maximum absolute value. (sklearn.preprocessing.MaxAbsScaler)

ScatterND ¶

Scatters updates into a tensor of shape equal to shape attribute according indices (TFLite),The output of the ScatterND layer is produced by creating a copy of the input data, and then updating its values to values specified by updates at specific index positions specified by indices (ONNX)

category: activation function
input data types: float32
output data types: float32

Specific constraints/recommendations:

5D or more indices (BATCH is not included) are not implemented, 5D or more data (BATCH is not included) are not implemented

Selu ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Shape ¶

Returns a tensor containing the shape of the input tensor

category: Reshaping operation
input data types: float32, int8, uint8
output data types: int32

Specific constraints/recommendations:

Shape layer is not supported as the only layer in a network;Shape layer only supported with know shapes in input; when batch size is undetermined, it is set to 1
end attribute is not supported
The attribute start different from the default value is not supported

Sigmoid ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Sign ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Sin ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Sinh ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Size ¶

Returns a tensor containing the number of elements of the input tensor

category: Reshaping operation
input data types: float32, int8, uint8
output data types: int32

Specific constraints/recommendations:

Size layer is not supported as the only layer in a network;Suze layer only supported with know shapes in input; when batch size is undetermined, it is set to 1
end attribute is not supported
The attribute start different from the default value is not supported

Slice ¶

Crops the input

category: reshaping layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Softmax ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Specific constraints/recommendations:

It is supported only for 1D tensor and only on the channel dimension
The value 1 is supported as the default value of the axis attribute, not the value -1

Softplus ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Softsign ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

SpaceToDepth ¶

Rearranges blocks of spatial data into depth

category: reshaping layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Split ¶

Splits a tensor into a list of sub tensors

category: split operator
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Specific constraints/recommendations:

Only supported if the number of splits is equal to the size of the splitting dimension

Sqrt ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Squeeze ¶

Reshapes a tensor

category: Reshaping operation
input data types: float32
output data types: float32

Sub ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

Sum ¶

Performs element-wise operation

category: eltwise operator
input data types: float32
output data types: float32

SVMClassifier ¶

SVMClassifier layer

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

used to support for example the skit-learn algo:
- C-Support Vector Classification. (sklearn.svm.SVC)
- Nu-Support Vector Classification (sklearn.svm.NuSVC)

SVMRegressor ¶

Support Vector Machine regression prediction and one-class SVM anomaly detection

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

kernel_type argument supports one of the following ‘LINEAR’,‘POLY’,‘RBF’,‘SIGMOID’.
used to support for example the skit-learn algo:
- Unsupervised Outlier Detection. (sklearn.svm.OneClassSVM)

Tan ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Tanh ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

ThresholdedRelu ¶

Applies an activation function to the input tensor

category: activation layer
input data types: float32
output data types: float32

Tile ¶

Constructs a tensor by tiling the input tensor

category: reshaping layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Specific constraints/recommendations:

tiling on batch-dimension is not supported

TopK ¶

Retrieve the top-K largest or smallest elements along a specified axis

category: topK operator
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Transpose ¶

Permutes the dimensions of the input according to a given pattern

category: reshaping layer
input data types: float32, int8, uint8
output data types: float32, int8, uint8

Specific constraints/recommendations:

transposing the batch dimension is not supported
perm attribute must have a length equal to the rank of the input

TreeEnsembleClassifier ¶

Tree ensemble classifier. Returns the top class for each of N inputs.

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

This layer supports compression levels.
- none: disable compresion and optimization
- lossless (default): enable optimization only
- low, medium, high: enable optimization and weight compression
used to support for example the skit-learn algo:
- A random forest classifier (sklearn.ensemble.RandomForestClassifier)
- An extra-trees classifier (sklearn.ensemble.ExtraTreesClassifier)
- Gradient Boosting for classification. (sklearn.ensemble.GradientBoostingClassifier)
- Histogram-based Gradient Boosting Classification Tree. (sklearn.ensemble.HistGradientBoostingClassifier)
- A decision tree classifier. (sklearn.tree.DecisionTreeClassifier)
For detailed information, see [Machine Learning support (ONNX-ML operators) [ONNX-ML]][STAI_CORE_ONNX_ML] article

TreeEnsembleRegressor ¶

Tree Ensemble regressor. Returns the regressed values for each input in N.

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

Warning: This operator is not supported for C code generation, it is only used during iForest generation. Not all arguments are supported.
used to support for example the skit-learn algo:
- A Bagging classifier (sklearn.ensemble.BaggingClassifier)
- An AdaBoost classifier (sklearn.ensemble.AdaBoostClassifier)
- A Bagging regressor (sklearn.ensemble.BaggingRegressor)
- A random forest regressor (sklearn.ensemble.RandomForestRegressor)
- An extra-trees regressor (sklearn.ensemble.ExtraTreesRegressor)
- Gradient Boosting for regression (sklearn.ensemble.GradientBoostingRegressor)
- Histogram-based Gradient Boosting Regression Tree (sklearn.ensemble.HistGradientBoostingRegressor)
- Isolation Forest Algorithm (sklearn.ensemble.IsolationForest)
For detailed information, see [Machine Learning support (ONNX-ML operators) [ONNX-ML]][STAI_CORE_ONNX_ML] article

Unsqueeze ¶

Reshapes a tensor

category: Reshaping operation
input data types: float32
output data types: float32

Upsample ¶

Resize the input tensor: it calculates every value in the output tensor as a weighted average of neighborhood

category: resizing operation
input data types: float32
output data types: float32

Specific constraints/recommendations:

Cubic resize is not supported
half_pixel_symmetric value for coordinate_tansformation_mode is not supported
The attribute antialias different from the default value is not supported
axes attribute is not supported
The attribute keep_aspect_ratio_policy different from the default value is not supported

Where ¶

Where layer

category: generic layer
input data types: float32, int8, uint8, int16, uint16, int32, uint32, bool
output data types: float32, int8, uint8, int16, uint16, int32, uint32, bool

Xor ¶

Performs boolean element-wise operation

category: eltwise operator
input data types: bool
output data types: bool

ZipMap ¶

Creates a map from the input and the attributes.

category: onnx.ml
input data types: float32
output data types: float32

Specific constraints/recommendations:

used to support for example the skit-learn algo:
- A decision tree classifier. (sklearn.tree.DecisionTreeClassifier)
- …