Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Implementation:Kserve Kserve ExplainerWrapper Base

From Leeroopedia
Revision as of 13:08, 16 February 2026 by Admin (talk | contribs) (Auto-imported from implementations/Kserve_Kserve_ExplainerWrapper_Base.md)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Knowledge Sources
Domains Explainability, Design Patterns
Last Updated 2026-02-13 00:00 GMT

Overview

Concrete tool for defining the abstract base interface that all Alibi explanation method wrappers must implement provided by the KServe sample code.

Description

ExplainerWrapper is the abstract base class for all Alibi-based explainer wrappers in the KServe Alibi explainer sample. It defines two methods: validate() which accepts an optional training data URL and performs validation (default is a no-op), and explain() which accepts a list of input instances and optional headers and returns an Alibi Explanation object. Subclasses such as AnchorImages, AnchorTabular, and AnchorText must override the explain() method with their specific explanation logic.

Usage

Use this class as a base when implementing a new Alibi explanation method wrapper. All concrete explainer wrappers should inherit from ExplainerWrapper and implement the explain() method.

Code Reference

Source Location

Signature

class ExplainerWrapper:
    def validate(self, training_data_url: Optional[str]):
        pass

    def explain(self, inputs: List, headers: Dict[str, str] = None) -> Explanation:
        pass

Import

from alibiexplainer.explainer_wrapper import ExplainerWrapper

I/O Contract

Inputs

validate()

Name Type Required Description
training_data_url Optional[str] No URL to training data for validation purposes

explain()

Name Type Required Description
inputs List Yes List of input instances to explain
headers Dict[str, str] No Optional HTTP headers from the request

Outputs

validate()

Name Type Description
(none) None No-op in the base class

explain()

Name Type Description
explanation Explanation Alibi Explanation object (no-op in the base class)

Usage Examples

Basic Usage

from alibiexplainer.explainer_wrapper import ExplainerWrapper
from alibi.api.interfaces import Explanation
from typing import List, Dict


class MyCustomExplainer(ExplainerWrapper):
    def explain(self, inputs: List, headers: Dict[str, str] = None) -> Explanation:
        # Custom explanation logic here
        ...
        return explanation

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment