Practical overview
When it fits
Use AIX360 when a machine-learning project needs structured model or data explanations and the appropriate method depends on factors such as data type, local versus global scope, or direct versus post-hoc interpretation.
Who it is for
- Python developers building explainability into machine-learning applications
- Data scientists investigating model behavior and data
- Researchers and teams creating trusted-ML evaluation or diagnostic workflows
Documented uses
- Analyse data and model behavior through multiple explanation approaches
- Produce local or global explanations for machine-learning models
- Evaluate explanations with documented faithfulness and monotonicity metrics
- Explore explainability techniques through tutorials, notebooks and an interactive experience
Requirements & limits
The public repository identifies AIX360 as open source under the Apache-2.0 license. Documented explainer configurations list macOS, Ubuntu and Windows support, with Python requirements varying by explainer.
- Select the method according to whether the task concerns data or models, and local or global explanations.
- Check the individual explainer's operating-system and Python-version requirements before setup.
- Use the taxonomy and guidance materials when choosing among the available algorithms.
- The repository notes that LLM explainability belongs to the separately identified ICX360 toolkit, not an established AIX360 capability.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Set up a selected AIX360 explainer
Prepare a documented AIX360 explainer configuration for local development.
Before you start
- Python version and operating system supported by the selected explainer.
- AIX360 repository or package installation materials.
Steps
Choose an explainer from the supported-configuration table.
Check its listed operating-system and Python-version requirements.
Follow the repository's setup instructions for that installation keyword.
Use the corresponding tutorials, example notebooks or API documentation to continue.
Expected outcome
A compatible local configuration for the selected explainer, followed by the documented learning and API materials.
Guide limitations
- Support varies by explainer; the repository lists different Python versions and operating-system combinations.
- This guide covers setup and next-stage documentation, not a model-explanation execution procedure.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 30 September 2026
- Original website title
- Trusted-AI/AIX360 · Official homepage title · retained in its source language