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

Example 31 with PerturbationContext

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

the class DataUtilsTest method testPerturbFeaturesEmpty.

@Test
void testPerturbFeaturesEmpty() {
    List<Feature> features = new LinkedList<>();
    PerturbationContext perturbationContext = new PerturbationContext(random, 0);
    List<Feature> newFeatures = DataUtils.perturbFeatures(features, perturbationContext);
    assertNotNull(newFeatures);
    assertEquals(features.size(), newFeatures.size());
}

Example 32 with PerturbationContext

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

the class DataUtilsTest method assertPerturbDropNumeric.

private void assertPerturbDropNumeric(PredictionInput input, int noOfPerturbations) {
    List<Feature> newFeatures = DataUtils.perturbFeatures(input.getFeatures(), new PerturbationContext(random, noOfPerturbations));
    int changedFeatures = 0;
    for (int i = 0; i < input.getFeatures().size(); i++) {
        double v = input.getFeatures().get(i).getValue().asNumber();
        double pv = newFeatures.get(i).getValue().asNumber();
        if (v != pv) {
            changedFeatures++;
        }
    }
    assertThat(changedFeatures).isBetween((int) Math.min(noOfPerturbations, input.getFeatures().size() * 0.5), (int) Math.max(noOfPerturbations, input.getFeatures().size() * 0.5));
}

Example 33 with PerturbationContext

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

the class PmmlScorecardCategoricalLimeExplainerTest method testExplanationImpactScoreWithOptimization.

@Test
void testExplanationImpactScoreWithOptimization() throws ExecutionException, InterruptedException {
    PredictionProvider model = getModel();
    List<PredictionInput> samples = getSamples();
    List<PredictionOutput> predictionOutputs = model.predictAsync(samples.subList(0, 5)).get();
    List<Prediction> predictions = DataUtils.getPredictions(samples, predictionOutputs);
    long seed = 0;
    LimeConfigOptimizer limeConfigOptimizer = new LimeConfigOptimizer().withDeterministicExecution(true).forImpactScore();
    Random random = new Random();
    PerturbationContext perturbationContext = new PerturbationContext(seed, random, 1);
    LimeConfig initialConfig = new LimeConfig().withSamples(10).withPerturbationContext(perturbationContext);
    LimeConfig optimizedConfig = limeConfigOptimizer.optimize(initialConfig, predictions, model);
    assertThat(optimizedConfig).isNotSameAs(initialConfig);
}

Example 34 with PerturbationContext

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

the class PmmlScorecardCategoricalLimeExplainerTest method testExplanationWeightedStabilityWithOptimization.

@Test
void testExplanationWeightedStabilityWithOptimization() throws ExecutionException, InterruptedException, TimeoutException {
    PredictionProvider model = getModel();
    List<PredictionInput> samples = getSamples();
    List<PredictionOutput> predictionOutputs = model.predictAsync(samples.subList(0, 5)).get();
    List<Prediction> predictions = DataUtils.getPredictions(samples, predictionOutputs);
    long seed = 0;
    LimeConfigOptimizer limeConfigOptimizer = new LimeConfigOptimizer().withDeterministicExecution(true).withWeightedStability(0.4, 0.6);
    Random random = new Random();
    PerturbationContext perturbationContext = new PerturbationContext(seed, random, 1);
    LimeConfig initialConfig = new LimeConfig().withSamples(10).withPerturbationContext(perturbationContext);
    LimeConfig optimizedConfig = limeConfigOptimizer.optimize(initialConfig, predictions, model);
    assertThat(optimizedConfig).isNotSameAs(initialConfig);
    LimeExplainer limeExplainer = new LimeExplainer(optimizedConfig);
    PredictionInput testPredictionInput = getTestInput();
    List<PredictionOutput> testPredictionOutputs = model.predictAsync(List.of(testPredictionInput)).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    Prediction instance = new SimplePrediction(testPredictionInput, testPredictionOutputs.get(0));
    assertDoesNotThrow(() -> ValidationUtils.validateLocalSaliencyStability(model, instance, limeExplainer, 1, 0.5, 0.7));
}

Example 35 with PerturbationContext

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

the class PmmlScorecardCategoricalLimeExplainerTest method testPMMLScorecardCategorical.

@Test
void testPMMLScorecardCategorical() throws Exception {
    PredictionInput input = getTestInput();
    Random random = new Random();
    LimeConfig limeConfig = new LimeConfig().withSamples(10).withPerturbationContext(new PerturbationContext(0L, random, 1));
    LimeExplainer limeExplainer = new LimeExplainer(limeConfig);
    PredictionProvider model = getModel();
    List<PredictionOutput> predictionOutputs = model.predictAsync(List.of(input)).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    assertThat(predictionOutputs).isNotNull().isNotEmpty();
    PredictionOutput output = predictionOutputs.get(0);
    assertThat(output).isNotNull();
    Prediction prediction = new SimplePrediction(input, output);
    Map<String, Saliency> saliencyMap = limeExplainer.explainAsync(prediction, model).get(Config.INSTANCE.getAsyncTimeout(), Config.INSTANCE.getAsyncTimeUnit());
    for (Saliency saliency : saliencyMap.values()) {
        assertThat(saliency).isNotNull();
        double v = ExplainabilityMetrics.impactScore(model, prediction, saliency.getTopFeatures(2));
        assertThat(v).isGreaterThan(0d);
    }
    assertDoesNotThrow(() -> ValidationUtils.validateLocalSaliencyStability(model, prediction, limeExplainer, 1, 0.4, 0.4));
    List<PredictionInput> inputs = getSamples();
    DataDistribution distribution = new PredictionInputsDataDistribution(inputs);
    String decision = "score";
    int k = 1;
    int chunkSize = 2;
    double f1 = ExplainabilityMetrics.getLocalSaliencyF1(decision, model, limeExplainer, distribution, k, chunkSize);
    AssertionsForClassTypes.assertThat(f1).isBetween(0d, 1d);
}