Package ej.microai

Class Tensor

java.lang.Object

ej.microai.Tensor

Direct Known Subclasses:

InputTensor, OutputTensor

public class Tensor
extends java.lang.Object

The Tensor class offers services to deal with MicroAI tensors.

Tensor are the input/output of your MLInferenceEngine.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	Tensor.DataType	The Tensor.DataType class enumerates the MicroAI data types.
static class	Tensor.QuantizationParameters	The QuantizationParameters class represents the quantization parameters used by a MLInferenceEngine.

Method Summary

Modifier and Type	Method	Description
int	getDataType()	Gets tensor data type.
int	getNumberBytes()	Gets the number of bytes of the tensor.
int	getNumberDimensions()	Gets the number of dimensions of the tensor.
int	getNumberElements()	Gets the number of elements of the tensor.
Tensor.QuantizationParameters	getQuantizationParams()	Gets parameters of asymmetric quantization for the tensor.
void	getShape(int[] sizes)	Gets the tensor shape.
boolean	isQuantized()	Gets the quantization status of the tensor.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

getDataType

public int getDataType()

Gets tensor data type.

Returns:

the tensor data type (see Tensor.DataType).

getNumberBytes

public int getNumberBytes()

Gets the number of bytes of the tensor.

Returns:

number of bytes of the tensor.

getNumberDimensions

public int getNumberDimensions()

Gets the number of dimensions of the tensor.

Returns:

number of dimensions of the tensor.

getNumberElements

public int getNumberElements()

Gets the number of elements of the tensor.

Returns:

number of elements of the tensor.

getQuantizationParams

public Tensor.QuantizationParameters getQuantizationParams()

Gets parameters of asymmetric quantization for the tensor.

Real values can be quantized using: quantized_value = real_value/scale + zero_point.

Returns:

the quantization parameters (see Tensor.QuantizationParameters).

getShape

public void getShape(int[] sizes)

Gets the tensor shape. Fill an array with the size of each dimension where the n-th element of the array correspond to the n-th dimension of the tensor.

Parameters:

sizes - is an array that contains the size of each dimension of the tensor.

isQuantized

public boolean isQuantized()

Gets the quantization status of the tensor.

Returns:

true if the tensor is quantized, false otherwise.