Explainability

Learn how to implement model explainability in Seldon Core using Alibi-Explain integration for black box model explanations and pipeline insights.

Explainers are Model resources with some extra settings. They allow a range of explainers from the Alibi-Explain library to be run on MLServer.

An example Anchors explainer definitions is shown below.

# samples/models/income-explainer.yaml
apiVersion: mlops.seldon.io/v1alpha1
kind: Model
metadata:
  name: income-explainer
spec:
  storageUri: "gs://seldon-models/scv2/samples/mlserver_1.5.0/income-sklearn/anchor-explainer"
  explainer:
    type: anchor_tabular
    modelRef: income

The key additions are:

• type: This must be one of thesupported Alibi Explainer types supported by the Alibi Explain runtime in MLServer.

• modelRef: The model name for black box explainers.

• pipelineRef: The pipeline name for black box explainers.

Only one of modelRef and pipelineRef is allowed.

Pipeline Explanations

Blackbox explainers can explain a Pipeline as well as a model. An example from the Huggingface sentiment demo is show below.

# samples/models/hf-sentiment-explainer.yaml
apiVersion: mlops.seldon.io/v1alpha1
kind: Model
metadata:
  name: sentiment-explainer
spec:
  storageUri: "gs://seldon-models/scv2/examples/huggingface/speech-sentiment/explainer"
  explainer:
    type: anchor_text
    pipelineRef: sentiment-explain

Examples

• Tabular income classification model with Anchor Tabular black box model explainer

• Huggingface Sentiment model with Anchor Text black box pipeline explainer

• Anchor Text movies sentiment explainer