Heroku LangChain.js - v1.0.1
    Preparing search index...

    Interface HerokuEmbeddingsCallOptions

    Interface for call-time options when using HerokuEmbeddings. These options can be passed to embedDocuments() and embedQuery() methods.

    interface HerokuEmbeddingsCallOptions {
        input_type?:
            | "search_document"
            | "search_query"
            | "classification"
            | "clustering";
        model?: string;
        encoding_format?: "base64"
        | "raw";
        embedding_type?: "float" | "int8" | "uint8" | "binary" | "ubinary";
        truncate?: "NONE" | "START" | "END";
        additionalKwargs?: Record<string, any>;
    }

    Hierarchy

    • RunnableConfig
      • HerokuEmbeddingsCallOptions
    Index

    Properties

    input_type? model? encoding_format? embedding_type? truncate? additionalKwargs?

    Properties

    input_type?:
        | "search_document"
        | "search_query"
        | "classification"
        | "clustering"

    Type of input text for embedding optimization.

    • "search_document": For storing documents in a search index
    • "search_query": For searching with a query
    • "classification": For classification tasks
    • "clustering": For clustering tasks
    model?: string

    Override the default model for this specific call.

    encoding_format?: "base64" | "raw"

    Format for the embeddings encoding.

    embedding_type?: "float" | "int8" | "uint8" | "binary" | "ubinary"

    Type of embedding to generate.

    truncate?: "NONE" | "START" | "END"

    How to handle inputs longer than the maximum length.

    additionalKwargs?: Record<string, any>

    Additional options to pass to the Heroku embeddings API.

    Settings

    Member Visibility

    On This Page

    Properties