org.apache.ignite.ml.selection.scoring.evaluator

Class Evaluator

java.lang.Object

org.apache.ignite.ml.selection.scoring.evaluator.Evaluator

public class Evaluator
extends Object

Evaluator that computes metrics from predictions and ground truth values.

Constructor Summary

Constructors

Constructor and Description
Evaluator()

Method Summary

Modifier and Type	Method and Description
static <K,V> EvaluationResult	evaluate(DatasetBuilder<K,V> datasetBuilder, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, Metric... metrics) Evaluate model.
static <K,V> double	evaluate(IgniteCache<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, Metric metric) Computes the given metric on the given cache.
static <K,V> double	evaluate(IgniteCache<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, MetricName metric) Computes the given metric on the given cache.
static <K,V> double	evaluate(IgniteCache<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, Metric metric) Computes the given metric on the given cache.
static <K,V> double	evaluate(IgniteCache<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, MetricName metric) Computes the given metric on the given cache.
static <K,V> double	evaluate(Map<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, Metric metric) Computes the given metric on the given cache.
static <K,V> double	evaluate(Map<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, MetricName metric) Computes the given metric on the given cache.
static <K,V> double	evaluate(Map<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, Metric metric) Computes the given metric on the given cache.
static <K,V> double	evaluate(Map<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor, MetricName metric) Computes the given metric on the given cache.
static <K,V> EvaluationResult	evaluateBinaryClassification(IgniteCache<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate binary classifier by default metrics (see package classification).
static <K,V> EvaluationResult	evaluateBinaryClassification(IgniteCache<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate binary classifier by default metrics (see package classification).
static <K,V> EvaluationResult	evaluateBinaryClassification(Map<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate binary classifier by default metrics (see package classification).
static <K,V> EvaluationResult	evaluateBinaryClassification(Map<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate binary classifier by default metrics (see package classification).
static <K,V> EvaluationResult	evaluateRegression(IgniteCache<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate regression by default metrics (see package regression).
static <K,V> EvaluationResult	evaluateRegression(IgniteCache<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate regression by default metrics (see package regression).
static <K,V> EvaluationResult	evaluateRegression(Map<K,V> dataCache, IgniteBiPredicate<K,V> filter, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate regression by default metrics (see package regression).
static <K,V> EvaluationResult	evaluateRegression(Map<K,V> dataCache, IgniteModel<Vector,Double> mdl, Preprocessor<K,V> preprocessor) Evaluate regression by default metrics (see package regression).

Methods inherited from class java.lang.Object

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

Constructor Detail

Evaluator

public Evaluator()

Method Detail

evaluate

public static <K,V> double evaluate(IgniteCache<K,V> dataCache,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    Metric metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

mdl - The model.

preprocessor - The preprocessor.

metric - The metric.

Returns:

Computed metric.

evaluate

public static <K,V> double evaluate(IgniteCache<K,V> dataCache,
                                    IgniteBiPredicate<K,V> filter,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    Metric metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The filter.

mdl - The model.

preprocessor - The preprocessor.

metric - The metric.

Returns:

Computed metric.

evaluate

public static <K,V> double evaluate(IgniteCache<K,V> dataCache,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    MetricName metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

mdl - The model.

preprocessor - The preprocessor.

metric - The metric name.

Returns:

Computed metric.

evaluate

public static <K,V> double evaluate(IgniteCache<K,V> dataCache,
                                    IgniteBiPredicate<K,V> filter,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    MetricName metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The given filter.

mdl - The model.

preprocessor - The preprocessor.

metric - The metric name.

Returns:

Computed metric.

evaluate

public static <K,V> double evaluate(Map<K,V> dataCache,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    Metric metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given local data.

mdl - The model.

preprocessor - The preprocessor.

metric - The binary classification metric.

Returns:

Computed metric.

evaluate

public static <K,V> double evaluate(Map<K,V> dataCache,
                                    IgniteBiPredicate<K,V> filter,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    Metric metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The given filter.

mdl - The model.

preprocessor - The preprocessor.

metric - The binary classification metric.

Returns:

Computed metric.

evaluate

public static <K,V> double evaluate(Map<K,V> dataCache,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    MetricName metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given local data.

mdl - The model.

preprocessor - The preprocessor.

metric - The binary classification metric.

Returns:

Computed metric.

evaluate

public static <K,V> double evaluate(Map<K,V> dataCache,
                                    IgniteBiPredicate<K,V> filter,
                                    IgniteModel<Vector,Double> mdl,
                                    Preprocessor<K,V> preprocessor,
                                    MetricName metric)

Computes the given metric on the given cache.

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The given filter.

mdl - The model.

preprocessor - The preprocessor.

metric - The binary classification metric.

Returns:

Computed metric.

evaluateBinaryClassification

public static <K,V> EvaluationResult evaluateBinaryClassification(IgniteCache<K,V> dataCache,
                                                                  IgniteModel<Vector,Double> mdl,
                                                                  Preprocessor<K,V> preprocessor)

Evaluate binary classifier by default metrics (see package classification). TODO: IGNITE-12156

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluateBinaryClassification

public static <K,V> EvaluationResult evaluateBinaryClassification(IgniteCache<K,V> dataCache,
                                                                  IgniteBiPredicate<K,V> filter,
                                                                  IgniteModel<Vector,Double> mdl,
                                                                  Preprocessor<K,V> preprocessor)

Evaluate binary classifier by default metrics (see package classification).

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The filter.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluateBinaryClassification

public static <K,V> EvaluationResult evaluateBinaryClassification(Map<K,V> dataCache,
                                                                  IgniteModel<Vector,Double> mdl,
                                                                  Preprocessor<K,V> preprocessor)

Evaluate binary classifier by default metrics (see package classification).

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluateBinaryClassification

public static <K,V> EvaluationResult evaluateBinaryClassification(Map<K,V> dataCache,
                                                                  IgniteBiPredicate<K,V> filter,
                                                                  IgniteModel<Vector,Double> mdl,
                                                                  Preprocessor<K,V> preprocessor)

Evaluate binary classifier by default metrics (see package classification).

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The given filter.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluateRegression

public static <K,V> EvaluationResult evaluateRegression(IgniteCache<K,V> dataCache,
                                                        IgniteModel<Vector,Double> mdl,
                                                        Preprocessor<K,V> preprocessor)

Evaluate regression by default metrics (see package regression).

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluateRegression

public static <K,V> EvaluationResult evaluateRegression(IgniteCache<K,V> dataCache,
                                                        IgniteBiPredicate<K,V> filter,
                                                        IgniteModel<Vector,Double> mdl,
                                                        Preprocessor<K,V> preprocessor)

Evaluate regression by default metrics (see package regression).

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The given filter.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluateRegression

public static <K,V> EvaluationResult evaluateRegression(Map<K,V> dataCache,
                                                        IgniteModel<Vector,Double> mdl,
                                                        Preprocessor<K,V> preprocessor)

Evaluate regression by default metrics (see package regression).

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluateRegression

public static <K,V> EvaluationResult evaluateRegression(Map<K,V> dataCache,
                                                        IgniteBiPredicate<K,V> filter,
                                                        IgniteModel<Vector,Double> mdl,
                                                        Preprocessor<K,V> preprocessor)

Evaluate regression by default metrics (see package regression).

Type Parameters:

K - The type of cache entry key.

V - The type of cache entry value.

Parameters:

dataCache - The given cache.

filter - The given filter.

mdl - The model.

preprocessor - The preprocessor.

Returns:

Computed metrics.

evaluate

public static <K,V> EvaluationResult evaluate(DatasetBuilder<K,V> datasetBuilder,
                                              IgniteModel<Vector,Double> mdl,
                                              Preprocessor<K,V> preprocessor,
                                              Metric... metrics)

Evaluate model.

Type Parameters:

K - Type of key.

V - Type of value.

Parameters:

mdl - The model.

datasetBuilder - Dataset builder.

preprocessor - Preprocessor.

metrics - Metrics to compute.

Returns:

Evaluation result.