Examples with DataDistribution org.kie.kogito.explainability.model.DataDistribution used on opensource projects

Example 16 with DataDistribution

use of org.kie.kogito.explainability.model.DataDistribution in project kogito-apps by kiegroup.

the class LimeExplainerTest method testWithDataDistribution.

@Test
void testWithDataDistribution() throws InterruptedException, ExecutionException, TimeoutException {
    Random random = new Random();
    PerturbationContext perturbationContext = new PerturbationContext(4L, random, 1);
    List<FeatureDistribution> featureDistributions = new ArrayList<>();
    int nf = 4;
    List<Feature> features = new ArrayList<>();
    for (int i = 0; i < nf; i++) {
        Feature numericalFeature = FeatureFactory.newNumericalFeature("f-" + i, Double.NaN);
        features.add(numericalFeature);
        List<Value> values = new ArrayList<>();
        for (int r = 0; r < 4; r++) {
            values.add(Type.NUMBER.randomValue(perturbationContext));
        }
        featureDistributions.add(new GenericFeatureDistribution(numericalFeature, values));
    }
    DataDistribution dataDistribution = new IndependentFeaturesDataDistribution(featureDistributions);
    LimeConfig limeConfig = new LimeConfig().withDataDistribution(dataDistribution).withPerturbationContext(perturbationContext).withSamples(10);
    LimeExplainer limeExplainer = new LimeExplainer(limeConfig);
    PredictionInput input = new PredictionInput(features);
    PredictionProvider model = TestUtils.getSumThresholdModel(random.nextDouble(), random.nextDouble());
    PredictionOutput output = model.predictAsync(List.of(input)).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit()).get(0);
    Prediction prediction = new SimplePrediction(input, output);
    Map<String, Saliency> saliencyMap = limeExplainer.explainAsync(prediction, model).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    assertThat(saliencyMap).isNotNull();
    String decisionName = "inside";
    Saliency saliency = saliencyMap.get(decisionName);
    assertThat(saliency).isNotNull();
}

Example 17 with DataDistribution

use of org.kie.kogito.explainability.model.DataDistribution in project kogito-apps by kiegroup.

the class AggregatedLimeExplainerTest method testExplainWithPredictions.

@ParameterizedTest
@ValueSource(ints = { 0, 1, 2, 3, 4 })
void testExplainWithPredictions(int seed) throws ExecutionException, InterruptedException {
    Random random = new Random();
    random.setSeed(seed);
    PredictionProvider sumSkipModel = TestUtils.getSumSkipModel(1);
    DataDistribution dataDistribution = DataUtils.generateRandomDataDistribution(3, 100, random);
    List<PredictionInput> samples = dataDistribution.sample(10);
    List<PredictionOutput> predictionOutputs = sumSkipModel.predictAsync(samples).get();
    List<Prediction> predictions = DataUtils.getPredictions(samples, predictionOutputs);
    AggregatedLimeExplainer aggregatedLimeExplainer = new AggregatedLimeExplainer();
    Map<String, Saliency> explain = aggregatedLimeExplainer.explainFromPredictions(sumSkipModel, predictions).get();
    assertNotNull(explain);
    assertEquals(1, explain.size());
    assertTrue(explain.containsKey("sum-but1"));
    Saliency saliency = explain.get("sum-but1");
    assertNotNull(saliency);
    List<String> collect = saliency.getPositiveFeatures(2).stream().map(FeatureImportance::getFeature).map(Feature::getName).collect(Collectors.toList());
    // skipped feature should not appear in top two positive features
    assertFalse(collect.contains("f1"));
}

Example 18 with DataDistribution

use of org.kie.kogito.explainability.model.DataDistribution in project kogito-apps by kiegroup.

the class DummyModelsLimeExplainerTest method testFixedOutput.

@ParameterizedTest
@ValueSource(longs = { 0 })
void testFixedOutput(long seed) throws Exception {
    Random random = new Random();
    List<Feature> features = new LinkedList<>();
    features.add(FeatureFactory.newNumericalFeature("f1", 6));
    features.add(FeatureFactory.newNumericalFeature("f2", 3));
    features.add(FeatureFactory.newNumericalFeature("f3", 5));
    PredictionProvider model = TestUtils.getFixedOutputClassifier();
    PredictionInput input = new PredictionInput(features);
    List<PredictionOutput> outputs = model.predictAsync(List.of(input)).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    Prediction prediction = new SimplePrediction(input, outputs.get(0));
    LimeConfig limeConfig = new LimeConfig().withSamples(10).withPerturbationContext(new PerturbationContext(seed, random, 1));
    LimeExplainer limeExplainer = new LimeExplainer(limeConfig);
    Map<String, Saliency> saliencyMap = limeExplainer.explainAsync(prediction, model).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    for (Saliency saliency : saliencyMap.values()) {
        assertNotNull(saliency);
        List<FeatureImportance> topFeatures = saliency.getTopFeatures(3);
        assertEquals(3, topFeatures.size());
        for (FeatureImportance featureImportance : topFeatures) {
            assertEquals(0, featureImportance.getScore());
        }
        assertEquals(0d, ExplainabilityMetrics.impactScore(model, prediction, topFeatures));
    }
    int topK = 1;
    double minimumPositiveStabilityRate = 0.5;
    double minimumNegativeStabilityRate = 0.5;
    TestUtils.assertLimeStability(model, prediction, limeExplainer, topK, minimumPositiveStabilityRate, minimumNegativeStabilityRate);
    List<PredictionInput> inputs = new ArrayList<>();
    for (int i = 0; i < 100; i++) {
        List<Feature> fs = new LinkedList<>();
        fs.add(TestUtils.getMockedNumericFeature());
        fs.add(TestUtils.getMockedNumericFeature());
        fs.add(TestUtils.getMockedNumericFeature());
        inputs.add(new PredictionInput(fs));
    }
    DataDistribution distribution = new PredictionInputsDataDistribution(inputs);
    int k = 2;
    int chunkSize = 10;
    String decision = "class";
    double precision = ExplainabilityMetrics.getLocalSaliencyPrecision(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(precision).isEqualTo(1);
    double recall = ExplainabilityMetrics.getLocalSaliencyRecall(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(recall).isEqualTo(1);
    double f1 = ExplainabilityMetrics.getLocalSaliencyF1(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(f1).isEqualTo(1);
}

Example 19 with DataDistribution

use of org.kie.kogito.explainability.model.DataDistribution in project kogito-apps by kiegroup.

the class DummyModelsLimeExplainerTest method testMapOneFeatureToOutputClassification.

@ParameterizedTest
@ValueSource(longs = { 0 })
void testMapOneFeatureToOutputClassification(long seed) throws Exception {
    Random random = new Random();
    int idx = 1;
    List<Feature> features = new LinkedList<>();
    features.add(FeatureFactory.newNumericalFeature("f1", 1));
    features.add(FeatureFactory.newNumericalFeature("f2", 1));
    features.add(FeatureFactory.newNumericalFeature("f3", 3));
    PredictionInput input = new PredictionInput(features);
    PredictionProvider model = TestUtils.getEvenFeatureModel(idx);
    List<PredictionOutput> outputs = model.predictAsync(List.of(input)).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    Prediction prediction = new SimplePrediction(input, outputs.get(0));
    LimeConfig limeConfig = new LimeConfig().withSamples(100).withPerturbationContext(new PerturbationContext(seed, random, 2));
    LimeExplainer limeExplainer = new LimeExplainer(limeConfig);
    Map<String, Saliency> saliencyMap = limeExplainer.explainAsync(prediction, model).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    for (Saliency saliency : saliencyMap.values()) {
        assertNotNull(saliency);
        List<FeatureImportance> topFeatures = saliency.getTopFeatures(3);
        assertEquals(3, topFeatures.size());
        assertEquals(1d, ExplainabilityMetrics.impactScore(model, prediction, topFeatures));
    }
    double minimumPositiveStabilityRate = 0.5;
    double minimumNegativeStabilityRate = 0.5;
    int topK = 1;
    TestUtils.assertLimeStability(model, prediction, limeExplainer, topK, minimumPositiveStabilityRate, minimumNegativeStabilityRate);
    List<PredictionInput> inputs = new ArrayList<>();
    for (int i = 0; i < 100; i++) {
        List<Feature> fs = new LinkedList<>();
        fs.add(TestUtils.getMockedNumericFeature());
        fs.add(TestUtils.getMockedNumericFeature());
        fs.add(TestUtils.getMockedNumericFeature());
        inputs.add(new PredictionInput(fs));
    }
    DataDistribution distribution = new PredictionInputsDataDistribution(inputs);
    int k = 2;
    int chunkSize = 10;
    String decision = "feature-" + idx;
    double precision = ExplainabilityMetrics.getLocalSaliencyPrecision(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(precision).isEqualTo(1);
    double recall = ExplainabilityMetrics.getLocalSaliencyRecall(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(recall).isEqualTo(1);
    double f1 = ExplainabilityMetrics.getLocalSaliencyF1(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(f1).isEqualTo(1);
}

Example 20 with DataDistribution

use of org.kie.kogito.explainability.model.DataDistribution in project kogito-apps by kiegroup.

the class DummyModelsLimeExplainerTest method testUnusedFeatureRegression.

@ParameterizedTest
@ValueSource(longs = { 0 })
void testUnusedFeatureRegression(long seed) throws Exception {
    Random random = new Random();
    int idx = 2;
    List<Feature> features = new LinkedList<>();
    features.add(TestUtils.getMockedNumericFeature(100));
    features.add(TestUtils.getMockedNumericFeature(20));
    features.add(TestUtils.getMockedNumericFeature(10));
    PredictionProvider model = TestUtils.getSumSkipModel(idx);
    PredictionInput input = new PredictionInput(features);
    List<PredictionOutput> outputs = model.predictAsync(List.of(input)).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    Prediction prediction = new SimplePrediction(input, outputs.get(0));
    LimeConfig limeConfig = new LimeConfig().withSamples(10).withPerturbationContext(new PerturbationContext(seed, random, 1));
    LimeExplainer limeExplainer = new LimeExplainer(limeConfig);
    Map<String, Saliency> saliencyMap = limeExplainer.explainAsync(prediction, model).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    for (Saliency saliency : saliencyMap.values()) {
        assertNotNull(saliency);
        List<FeatureImportance> topFeatures = saliency.getTopFeatures(3);
        assertEquals(3, topFeatures.size());
        assertEquals(1d, ExplainabilityMetrics.impactScore(model, prediction, topFeatures));
    }
    int topK = 1;
    double minimumPositiveStabilityRate = 0.5;
    double minimumNegativeStabilityRate = 0.5;
    TestUtils.assertLimeStability(model, prediction, limeExplainer, topK, minimumPositiveStabilityRate, minimumNegativeStabilityRate);
    List<PredictionInput> inputs = new ArrayList<>();
    for (int i = 0; i < 100; i++) {
        List<Feature> fs = new LinkedList<>();
        fs.add(TestUtils.getMockedNumericFeature());
        fs.add(TestUtils.getMockedNumericFeature());
        fs.add(TestUtils.getMockedNumericFeature());
        inputs.add(new PredictionInput(fs));
    }
    DataDistribution distribution = new PredictionInputsDataDistribution(inputs);
    int k = 2;
    int chunkSize = 10;
    String decision = "sum-but" + idx;
    double precision = ExplainabilityMetrics.getLocalSaliencyPrecision(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(precision).isEqualTo(1);
    double recall = ExplainabilityMetrics.getLocalSaliencyRecall(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(recall).isEqualTo(1);
    double f1 = ExplainabilityMetrics.getLocalSaliencyF1(decision, model, limeExplainer, distribution, k, chunkSize);
    assertThat(f1).isEqualTo(1);
}