The alibi explanation API takes inspiration from scikit-learn, consisting of distinct initialize,įit and explain steps. Which can be installed to the base conda enviroment with: conda install mamba -n base -c conda-forgeįor the standard Alibi install: mamba install -c conda-forge alibiįor distributed computating support: mamba install -c conda-forge alibi rayįor SHAP support: mamba install -c conda-forge alibi shap To install from conda-forge it is recommended to use mamba, To take advantage of distributed computation of explanations, install alibi with ray: pip install alibi įor SHAP support, install alibi as follows: pip install alibi If you're interested in outlier detection, concept drift or adversarial instance detection, check out our sister project alibi-detect.Īlibi can be installed from PyPI: pip install alibiĪlternatively, the development version can be installed: pip install git+ The focus of the library is to provide high-quality implementations of black-box, white-box, local and globalĮxplanation methods for classification and regression models. Alibi is an open source Python library aimed at machine learning model inspection and interpretation.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |