1. What-If Tool Visually probe the behavior of trained machine learning models with minimal coding — explore model performance, fairness, and counterfactual examples interactively.
2. DeepLIFT Deep Learning Important FeaTures — assigns importance scores to input features by comparing the activation of each neuron to its reference activation.
3. LIME Local Interpretable Model-Agnostic Explanations — explains predictions of any machine learning classifier by learning a locally faithful interpretable model around each prediction.
4. SHAP SHapley Additive exPlanations — a game theoretic approach to explain the output of any machine learning model, connecting optimal credit allocation with local explanations.
5. Rulex Platform Machine learning capabilities that ensure essential transparency, peak performance, and unwavering accuracy throughout the entire forecasting process — with native XAI built in.
6. Activation Atlases Uses feature inversion to visualize millions of activations from an image classification network — creating an explorable atlas of features the network has learned to represent concepts.
7. Ethical Institute XAI Framework Introduces explainability and bias evaluation in AI systems using a tool-plus-process approach — going beyond algorithms to operationalize responsible AI practices.
8. ELI5 Python library for visualizing and debugging various machine learning models using a unified API — supports scikit-learn, XGBoost, LightGBM, and more.
9. InterpretML A toolkit to help understand models and enable responsible machine learning — with Explainable Boosting Machine (EBM) and support for SHAP, LIME, and other explainers.
10. OmniXAI Omni eXplainable AI Python library offering omni-way explainability and interpretable ML capabilities — addressing many pain points in explaining ML model decisions across tabular, text, image, and time-series data.
11. Shapash Python library making machine learning interpretable and accessible to everyone — with clear, explicit visualization labels that facilitate understanding for both technical and non-technical audiences.
12. TCAV Testing with Concept Activation Vectors — interpretability method to understand what signals neural network models use for prediction using human-friendly concepts.
13. DALEX X-rays any model and helps explore and explain its behaviour — providing consistent tools for understanding how complex black-box models produce their predictions.
14. Alibi Explain Open source Python library for machine learning model inspection and interpretation — providing high-quality black-box, white-box, local, and global explanation methods for classification and regression models.
15. ExplainerDashboard Library for quickly building interactive dashboards to analyze and explain ML model predictions and workings — supporting scikit-learn, XGBoost, CatBoost, and LightGBM models.