org.nd4j.linalg.api.ops.impl.summarystats

Class Variance

java.lang.Object

org.nd4j.autodiff.functions.DifferentialFunction

org.nd4j.linalg.api.ops.BaseOp

org.nd4j.linalg.api.ops.BaseReduceOp

org.nd4j.linalg.api.ops.impl.summarystats.Variance

All Implemented Interfaces:

Op, ReduceOp

Direct Known Subclasses:

StandardDeviation

public class Variance
extends BaseReduceOp

Nested Class Summary

Nested classes/interfaces inherited from interface org.nd4j.linalg.api.ops.Op

Op.Type

Field Summary

Fields

Modifier and Type	Field and Description
protected double	bias
protected boolean	biasCorrected
protected double	mean

Fields inherited from class org.nd4j.linalg.api.ops.BaseReduceOp

isComplex, isEmptyReduce, keepDims

Fields inherited from class org.nd4j.linalg.api.ops.BaseOp

dimensionz, extraArgz, x, xVertexId, y, yVertexId, z, zVertexId

Fields inherited from class org.nd4j.autodiff.functions.DifferentialFunction

dimensions, extraArgs, inPlace, ownName, ownNameSetWithDefault, sameDiff, scalarValue

Constructor Summary

Constructors

Constructor and Description
Variance()
Variance(boolean biasCorrected)
Variance(INDArray x, boolean biasCorrected, boolean keepDims, int... dimensions)
Variance(INDArray x, boolean biasCorrected, int... dimensions)
Variance(INDArray x, INDArray z, boolean biasCorrected, boolean keepDims, int... dimensions)
Variance(INDArray x, INDArray z, boolean biasCorrected, int... dimensions)
Variance(INDArray x, int... dimension)
Variance(SameDiff sameDiff, SDVariable i_v, boolean biasCorrected, boolean keepDims, int[] dimensions)

Method Summary

Modifier and Type	Method and Description
List<DataType>	calculateOutputDataTypes(List<DataType> dataTypes) Calculate the data types for the output arrays.
List<LongShapeDescriptor>	calculateOutputShape() Calculate the output shape for this op
List<LongShapeDescriptor>	calculateOutputShape(OpContext oc)
List<SDVariable>	doDiff(List<SDVariable> grad) The actual implementation for automatic differentiation.
Op.Type	getOpType()
boolean	isBiasCorrected()
INDArray	noOp() Returns the no op version of the input Basically when a reduce can't happen (eg: sum(0) on a row vector) you have a no op state for a given reduction.
String	onnxName() The opName of this function in onnx
String	opName() The name of the op
int	opNum() The number of the op (mainly for old legacy XYZ ops like Op)
Op.Type	opType() The type of the op
DataType	resultType() This method returns datatype for result array wrt given inputs
DataType	resultType(OpContext oc)
void	setBiasCorrected(boolean biasCorrected)
String	tensorflowName() The opName of this function tensorflow
boolean	validateDataTypes(OpContext oc)

Methods inherited from class org.nd4j.linalg.api.ops.BaseReduceOp

hasReductionIndices, initFromOnnx, initFromTensorFlow, isComplexAccumulation, isKeepDims, setDimensions

Methods inherited from class org.nd4j.linalg.api.ops.BaseOp

clearArrays, defineDimensions, dimensions, equals, extraArgs, extraArgsBuff, extraArgsDataBuff, getFinalResult, getInputArgument, getNumOutputs, getOpType, hashCode, outputVariables, setX, setY, setZ, toCustomOp, toString, x, y, z

Methods inherited from class org.nd4j.autodiff.functions.DifferentialFunction

arg, arg, argNames, args, attributeAdaptersForFunction, configFieldName, diff, dup, getValue, isConfigProperties, larg, mappingsForFunction, onnxNames, outputs, outputVariable, outputVariables, outputVariablesNames, propertiesForFunction, rarg, replaceArg, setInstanceId, setPropertiesForFunction, setValueFor, tensorflowNames

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface org.nd4j.linalg.api.ops.ReduceOp

dimensions, getFinalResult

Methods inherited from interface org.nd4j.linalg.api.ops.Op

clearArrays, extraArgs, extraArgsBuff, extraArgsDataBuff, setExtraArgs, setX, setY, setZ, toCustomOp, x, y, z

Field Detail

mean

protected double mean

bias

protected double bias

biasCorrected

protected boolean biasCorrected

Constructor Detail

Variance

public Variance(SameDiff sameDiff,
                SDVariable i_v,
                boolean biasCorrected,
                boolean keepDims,
                int[] dimensions)

Variance

public Variance()

Variance

public Variance(boolean biasCorrected)

Variance

public Variance(INDArray x,
                int... dimension)

Variance

public Variance(INDArray x,
                INDArray z,
                boolean biasCorrected,
                int... dimensions)

Variance

public Variance(INDArray x,
                boolean biasCorrected,
                boolean keepDims,
                int... dimensions)

Variance

public Variance(INDArray x,
                boolean biasCorrected,
                int... dimensions)

Variance

public Variance(INDArray x,
                INDArray z,
                boolean biasCorrected,
                boolean keepDims,
                int... dimensions)

Method Detail

noOp

public INDArray noOp()

Description copied from interface: ReduceOp

Returns the no op version of the input Basically when a reduce can't happen (eg: sum(0) on a row vector) you have a no op state for a given reduction. For most accumulations, this should return x but certain transformations should return say: the absolute value

Specified by:

noOp in interface ReduceOp

Overrides:

noOp in class BaseReduceOp

Returns:

the no op version of the input

opNum

public int opNum()

Description copied from class: DifferentialFunction

The number of the op (mainly for old legacy XYZ ops like Op)

Specified by:

opNum in interface Op

Overrides:

opNum in class DifferentialFunction

Returns:

opName

public String opName()

Description copied from class: DifferentialFunction

The name of the op

Specified by:

opName in interface Op

Overrides:

opName in class DifferentialFunction

Returns:

the opName of this operation

isBiasCorrected

public boolean isBiasCorrected()

setBiasCorrected

public void setBiasCorrected(boolean biasCorrected)

doDiff

public List<SDVariable> doDiff(List<SDVariable> grad)

Description copied from class: DifferentialFunction

The actual implementation for automatic differentiation.

Specified by:

doDiff in class DifferentialFunction

Returns:

onnxName

public String onnxName()

Description copied from class: DifferentialFunction

The opName of this function in onnx

Overrides:

onnxName in class BaseOp

Returns:

tensorflowName

public String tensorflowName()

Description copied from class: DifferentialFunction

The opName of this function tensorflow

Overrides:

tensorflowName in class BaseOp

Returns:

getOpType

public Op.Type getOpType()

resultType

public DataType resultType()

Description copied from interface: ReduceOp

This method returns datatype for result array wrt given inputs

Returns:

resultType

public DataType resultType(OpContext oc)

validateDataTypes

public boolean validateDataTypes(OpContext oc)

calculateOutputShape

public List<LongShapeDescriptor> calculateOutputShape()

Description copied from class: DifferentialFunction

Calculate the output shape for this op

Specified by:

calculateOutputShape in class BaseReduceOp

Returns:

List of output shape descriptors

calculateOutputShape

public List<LongShapeDescriptor> calculateOutputShape(OpContext oc)

Overrides:

calculateOutputShape in class DifferentialFunction

opType

public Op.Type opType()

Description copied from class: DifferentialFunction

The type of the op

Overrides:

opType in class DifferentialFunction

Returns:

calculateOutputDataTypes

public List<DataType> calculateOutputDataTypes(List<DataType> dataTypes)

Description copied from class: DifferentialFunction

Calculate the data types for the output arrays. Though datatypes can also be inferred from DifferentialFunction.calculateOutputShape(), this method differs in that it does not require the input arrays to be populated. This is important as it allows us to do greedy datatype inference for the entire net - even if arrays are not available.

Overrides:

calculateOutputDataTypes in class DifferentialFunction

Parameters:

dataTypes - The data types of the inputs

Returns:

The data types of the outputs