Class Mean<T extends TNumber>

java.lang.Object

org.tensorflow.framework.metrics.BaseMetric

org.tensorflow.framework.metrics.Mean<T>

Type Parameters:

T - The data type for the metric result

All Implemented Interfaces:

Metric

Direct Known Subclasses:

Accuracy, BinaryAccuracy, BinaryCrossentropy, CategoricalAccuracy, CategoricalCrossentropy, CategoricalHinge, CosineSimilarity, Hinge, KLDivergence, LogCoshError, MeanAbsoluteError, MeanAbsolutePercentageError, MeanRelativeError, MeanSquaredError, MeanSquaredLogarithmicError, Poisson, RootMeanSquaredError, SparseCategoricalAccuracy, SparseCategoricalCrossentropy, SparseTopKCategoricalAccuracy, SquaredHinge, TopKCategoricalAccuracy

public class Mean<T extends TNumber>
extends BaseMetric

A metric that that implements a weighted mean MetricReduction.WEIGHTED_MEAN

Field Summary

Fields

Modifier and Type	Field	Description
protected Variable<T>	count	the variable that holds the count of the metric values.
static final String	COUNT
protected final MetricReduction	reduction
protected Variable<T>	total	the variable that holds the total of the metric values
static final String	TOTAL

Constructor Summary

Constructors

Constructor	Description
Mean(long seed, Class<T> type)	Creates a Reducible Metric with a metric reductions of MetricReduction.SUM and using Class.getSimpleName() for the metric name.
Mean(String name, long seed, Class<T> type)	Creates a Reducible Metric with a metric reductions of MetricReduction.SUM

Method Summary

Modifier and Type	Method	Description
Variable<T>	getCount()	Gets the count variable
Class<T>	getInternalType()	Gets the type for the variables
Variable<T>	getTotal()	Gets the total variable
protected void	init(Ops tf)	Initialize the TensorFlow Ops
Op	resetStates(Ops tf)	Resets any state variables to their initial values
<U extends TNumber> Operand<U>	result(Ops tf, Class<U> type)	Gets the current result of the metric
List<Op>	updateStateList(Ops tf, Operand<? extends TNumber> values, Operand<? extends TNumber> sampleWeights)	Updates the metric variables based on the inputs.

Methods inherited from class BaseMetric

callOnce, checkIsGraph, getName, getSeed, getTF, getVariableName, isInitialized, setInitialized, setName, setTF, updateState, updateState, updateStateList

Modifier and Type	Method	Description
final <T extends TNumber> Operand<T>	callOnce(Ops tf, Operand<? extends TNumber> values, Operand<? extends TNumber> sampleWeights, Class<T> type)	Calls update state once, followed by a call to get the result
protected void	checkIsGraph(Ops tf)	Checks if the TensorFlow Ops encapsulates a Graph environment.
String	getName()	The name for this metric.
long	getSeed()	Gets the random number generator seed value
protected Ops	getTF()	Gets the TensorFlow Ops for this metric
protected String	getVariableName(String varName)	Gets a formatted name for a variable, in the form BaseMetric.name + "_" + varName.
boolean	isInitialized()	Checks whether the Metric is initialized or not.
protected void	setInitialized(boolean initialized)	Sets the initialized indicator
void	setName(String name)	Sets the metric name
protected void	setTF(Ops tf)	Sets the TensorFlow Ops for this metric.
final Op	updateState(Ops tf, Operand<? extends TNumber> values, Operand<? extends TNumber> sampleWeights)	Creates a NoOp Operation with control dependencies to update the metric state
final Op	updateState(Ops tf, Operand<? extends TNumber> labels, Operand<? extends TNumber> predictions, Operand<? extends TNumber> sampleWeights)	Creates a NoOp Operation with control dependencies to update the metric state
List<Op>	updateStateList(Ops tf, Operand<? extends TNumber> labels, Operand<? extends TNumber> predictions, Operand<? extends TNumber> sampleWeights)	Creates a List of Operations to update the metric state based on labels and predictions.

Methods inherited from class Object

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

Field Details

TOTAL

public static final String TOTAL

See Also:

• Constant Field Values

COUNT

public static final String COUNT

See Also:

• Constant Field Values

reduction

protected final MetricReduction reduction

total

protected Variable<T extends TNumber> total

the variable that holds the total of the metric values

count

protected Variable<T extends TNumber> count

the variable that holds the count of the metric values. For MetricReduction.WEIGHTED_MEAN, this count may be weighted

Constructor Details

Mean

public Mean(long seed,
 Class<T> type)

Creates a Reducible Metric with a metric reductions of MetricReduction.SUM and using Class.getSimpleName() for the metric name.

Parameters:

seed - the seed for random number generation. An initializer created with a given seed will always produce the same random tensor for a given shape and data type.

type - the type for the variables and result

Mean

public Mean(String name,
 long seed,
 Class<T> type)

Creates a Reducible Metric with a metric reductions of MetricReduction.SUM

Parameters:

name - the name for this metric. If null, name defaults to Class.getSimpleName().

seed - the seed for random number generation. An initializer created with a given seed will always produce the same random tensor for a given shape and data type.

type - the type for the variables and result

Method Details

init

protected void init(Ops tf)

Initialize the TensorFlow Ops

Specified by:

init in class BaseMetric

Parameters:

tf - the TensorFlow Ops encapsulating a Graph environment.

resetStates

public Op resetStates(Ops tf)

Resets any state variables to their initial values

Parameters:

tf - the TensorFlow Ops encapsulating a Graph environment.

Returns:

the operation for doing the reset

updateStateList

public List<Op> updateStateList(Ops tf,
 Operand<? extends TNumber> values,
 Operand<? extends TNumber> sampleWeights)

Updates the metric variables based on the inputs. At least one input arg required for values, an optional additional input for the sampleWeights

Specified by:

updateStateList in interface Metric

Overrides:

updateStateList in class BaseMetric

Parameters:

tf - the TensorFlow Ops encapsulating a Graph environment.

values - the inputs to be passed to update state, this may not be null

sampleWeights - sample weights to be applied to values, will default to 1 if null.

Returns:

the result with a control dependency on update state Operands

Throws:

IllegalArgumentException - if values is null

result

public <U extends TNumber> Operand<U> result(Ops tf,
 Class<U> type)

Gets the current result of the metric

Type Parameters:

U - the date type for the result

Parameters:

tf - the TensorFlow Ops encapsulating a Graph environment.

type - the data type for the result

Returns:

the result, possibly with control dependencies

getTotal

public Variable<T> getTotal()

Gets the total variable

Returns:

the total variable

getCount

public Variable<T> getCount()

Gets the count variable

Returns:

the count variable

getInternalType

public Class<T> getInternalType()

Gets the type for the variables

Returns:

the type for the variables