CosineSimilarity

public class CosineSimilarity

A metric that computes the cosine similarity metric between labels and predictions.

Constants

int DEFAULT_AXIS

Inherited Constants

Public Constructors

CosineSimilarity(Ops tf, String name, long seed, Class<T> type)
Creates a metric that computes the cosine similarity metric between labels and predictions with a default axis, DEFAULT_AXIS
CosineSimilarity(Ops tf, String name, int axis, long seed, Class<T> type)
Creates a metric that computes the cosine similarity metric between labels and predictions.
CosineSimilarity(Ops tf, String name, int[] axis, long seed, Class<T> type)
Creates a CosineSimilarity metric

Public Methods

Operand<T>
call(Operand<? extends TNumber> labels, Operand<? extends TNumber> predictions)
Calculates the weighted loss between labels and predictions

Inherited Methods

Constants

public static final int DEFAULT_AXIS

Constant Value: -1

Public Constructors

public CosineSimilarity (Ops tf, String name, long seed, Class<T> type)

Creates a metric that computes the cosine similarity metric between labels and predictions with a default axis, DEFAULT_AXIS

Parameters
tf the TensorFlow Ops
name the name of this metric, if null then metric name is 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

public CosineSimilarity (Ops tf, String name, int axis, long seed, Class<T> type)

Creates a metric that computes the cosine similarity metric between labels and predictions.

Parameters
tf the TensorFlow Ops
name the name of this metric, if null then metric name is getSimpleName().
axis The dimension along which the cosine similarity is computed.
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

public CosineSimilarity (Ops tf, String name, int[] axis, long seed, Class<T> type)

Creates a CosineSimilarity metric

Parameters
tf the TensorFlow Ops
name the name of this metric, if null then metric name is getSimpleName().
axis The dimension along which the cosine similarity is computed.
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

Public Methods

public Operand<T> call (Operand<? extends TNumber> labels, Operand<? extends TNumber> predictions)

Calculates the weighted loss between labels and predictions

Parameters
labels the truth values or labels
predictions the predictions
Returns
  • the loss