12 projects
aiverify-test-engine
AI Verify Test Engine provides core interfaces, converters, data, model and plugin managers to facilitate the development of tests for AI systems. It is used as a base library for all AI Verify official stock-plugins and can be used to develop custom plugins.
aiverify-veritastool
Veritas Diagnosis tool for fairness & transparency assessment.
aiverify-shap-toolbox
AI Verify SHAP Toolbox provides SHAP (SHapley Additive exPlanations) methods to explain the output of machine learning models.
aiverify-robustness-toolbox
AI Verify Robustness Toolbox generates a perturbed dataset using boundary attack algorithm on the test dataset.
aiverify-partial-dependence-plot
AI Verify implementation of Partial Dependence Plot (PDP) that explains how each feature and its feature value contribute to the predictions.
aiverify-general-corruptions
Part of AI Verify image corruption toolbox. This package includes algorithms that add general corruptions (gaussian, poisson and salt and pepper noise) to images at different severity levels, to test the robustness of machine learning models.
aiverify-environment-corruptions
Part of AI Verify image corruption toolbox. This package includes algorithms that add environmental corruptions (rain, fog and snow) to images at different severity levels, to test the robustness of machine learning models.
aiverify-digital-corruptions
Part of AI Verify image corruption toolbox. This package includes algorithms that add digital corruptions (brightness up and down, contrast up and down, compression, random tilt, saturate) to images at different severity levels, to test the robustness of machine learning models.
aiverify-blur-corruptions
Part of AI Verify image corruption toolbox. This package includes algorithms that adds blur corruptions (defocus, gaussian, glass, horizontal motion, vertical motion and zoom Blur) to images at different severity levels, to test the robustness of machine learning models.
aiverify-fairness-metrics-toolbox-for-regression
AI Verify Fairness Metrics Toolbox (FMT) for Regression contains a list of fairness metrics used to measure how resources (e.g. opportunities, food, loan, medical help) are allocated among the demographic groups (e.g. married male, married female) given a set of sensitive feature(s) (e.g. gender, marital status). This plugin is developed for regression models.
aiverify-fairness-metrics-toolbox-for-classification
AI Verify Fairness Metrics Toolbox (FMT) for Classification contains a list of fairness metrics to measure how resources (e.g. opportunities, food, loan, medical help) are allocated among the demographic groups (e.g. married male, married female) given a set of sensitive feature(s) (e.g. gender, marital status). This plugin is developed for classification models.
aiverify-accumulated-local-effect
AI Verify implementation of the Accumulated Local Effect algorithm. The algorithm provides black box explainations of how features and their corresponding values influence the prediction of a model.
