Databricks SDK for JavaScript
    Preparing search index...

    Interface AiClassifyFunction

    Configuration for a task that runs the ai_classify function.

    The label set comes from exactly one of two places: inline_labels, which holds up to 10 labels, or a Unity Catalog table named by label_table. Use label_table for a larger label set.

    interface AiClassifyFunction {
        enableConfidenceScores?: boolean;
        enableRationales?: boolean;
        instructions?: string;
        labelSource?: {
            $case: "inlineLabels";
            inlineLabels: AiClassifyFunction_InlineLabels;
        };
        multilabel?: boolean;
        version?: string;
    }
    Index

    Properties

    enableConfidenceScores?: boolean

    When true, includes a per-label confidence score in the response.

    enableRationales?: boolean

    When true, includes a rationale explaining each classification in the response.

    instructions?: string

    Global description of the task and domain, passed to the model to improve classification quality.

    labelSource?: {
        $case: "inlineLabels";
        inlineLabels: AiClassifyFunction_InlineLabels;
    }

    Where the labels come from. Exactly one source must be set.

    Type Declaration

    multilabel?: boolean

    When true, more than one label can be returned for each input.

    version?: string

    The ai_classify function version to invoke (for example, 2.1).