Model explanations allow you to understand why a machine learning model gave a particular predictions. Seldon Deploy allows you to use Black Box Modle Explainers that can be used on models built using any technology.
Seldon Deploy supports a subset of the methods currently available in Alibi for both Seldon Core and KFServing deployments. Find documentation on Anchor explanation algorithm and read about creating your custom explainer
| Explanation Method | Seldon Core | KFServing | Alibi Docs |
|---|---|---|---|
| Anchor Tabular | Example | Example | Docs |
| Anchor Text | Example | Example | Docs |
| Anchor Images | Example | Example | Docs |
Explainer model classifier with tabular features
Explain models whose input is textual
Explain image based classifiers