This repo is the main hub for TrustyAI, containing the core Java library as well as various modules to support the TrustyAI Service, TrustyAI Operator, and TrustyAI Python Library.
TrustyAI is, at its core, a Java library and service for Explainable AI (XAI). TrustyAI offers fairness metrics, explainable AI algorithms, and various other XAI tools at a library-level as well as a containerized service and Kubernetes deployment.
- explainability-core, the core TrustyAI Java module, containing fairness metrics, AI explainers, and other XAI utilities.
- explainability-service, TrustyAI-as-a-service, a REST service for fairness metrics and explainability algorithms including ModelMesh integration.
- explainability-arrow, a Java module to facilitate the communication between TrustyAI-Java and TrustyAI-Python using Arrow.
- explainability-connectors, A Java module to interface with different black-box predictive model deployments or inference services. Includes support for KServe / ModelMesh via gRPC.
- explainability-integrationtests A set of integration tests for integrations within Kogito, namely DMN, PMML, and OpenNLP models.
Our preprint TrustyAI Explainability Toolkit is a great source of knowledge of what the core library can offer.
Furthermore, you can reach the dev team on:
- ODH Community Slack
- Zulip
- or by coming to one of our community meetings
All contributions are welcome! Before you start please read the contribution guide.