org.tensorflow.op.core

Class Multinomial<U extends Number>

java.lang.Object

org.tensorflow.op.PrimitiveOp

org.tensorflow.op.core.Multinomial<U>

Type Parameters:

U - data type for output() output

All Implemented Interfaces:

Op, Operand<U>

@Operator
public final class Multinomial<U extends Number>
extends PrimitiveOp
implements Operand<U>

Draws samples from a multinomial distribution.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	Multinomial.Options Optional attributes for Multinomial

Field Summary

Fields inherited from class org.tensorflow.op.PrimitiveOp

operation

Method Summary

Modifier and Type	Method and Description
Output<U>	asOutput() Returns the symbolic handle of a tensor.
static <U extends Number,T extends Number> Multinomial<U>	create(Scope scope, Operand<T> logits, Operand<Integer> numSamples, Class<U> outputDtype, Multinomial.Options... options) Factory method to create a class to wrap a new Multinomial operation to the graph.
static <T extends Number> Multinomial<Long>	create(Scope scope, Operand<T> logits, Operand<Integer> numSamples, Multinomial.Options... options) Factory method to create a class to wrap a new Multinomial operation to the graph, using default output types.
Output<U>	output() 2-D Tensor with shape `[batch_size, num_samples]`.
static Multinomial.Options	seed(Long seed)
static Multinomial.Options	seed2(Long seed2)

Methods inherited from class org.tensorflow.op.PrimitiveOp

equals, hashCode, toString

Methods inherited from class java.lang.Object

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

Method Detail

create

public static <U extends Number,T extends Number> Multinomial<U> create(Scope scope,
                                                                        Operand<T> logits,
                                                                        Operand<Integer> numSamples,
                                                                        Class<U> outputDtype,
                                                                        Multinomial.Options... options)

Factory method to create a class to wrap a new Multinomial operation to the graph.

Parameters:

scope - current graph scope

logits - 2-D Tensor with shape `[batch_size, num_classes]`. Each slice `[i, :]` represents the unnormalized log probabilities for all classes.

numSamples - 0-D. Number of independent samples to draw for each row slice.

outputDtype -

options - carries optional attributes values

Returns:

a new instance of Multinomial

create

public static <T extends Number> Multinomial<Long> create(Scope scope,
                                                          Operand<T> logits,
                                                          Operand<Integer> numSamples,
                                                          Multinomial.Options... options)

Factory method to create a class to wrap a new Multinomial operation to the graph, using default output types.

Parameters:

scope - current graph scope

logits - 2-D Tensor with shape `[batch_size, num_classes]`. Each slice `[i, :]` represents the unnormalized log probabilities for all classes.

numSamples - 0-D. Number of independent samples to draw for each row slice.

options - carries optional attributes values

Returns:

a new instance of Multinomial

seed

public static Multinomial.Options seed(Long seed)

Parameters:

seed - If either seed or seed2 is set to be non-zero, the internal random number generator is seeded by the given seed. Otherwise, a random seed is used.

seed2

public static Multinomial.Options seed2(Long seed2)

Parameters:

seed2 - A second seed to avoid seed collision.

output

public Output<U> output()

2-D Tensor with shape `[batch_size, num_samples]`. Each slice `[i, :]` contains the drawn class labels with range `[0, num_classes)`.

asOutput

public Output<U> asOutput()

Description copied from interface: Operand

Returns the symbolic handle of a tensor.

Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.

Specified by:

asOutput in interface Operand<U extends Number>

See Also:

OperationBuilder.addInput(Output)