Class OllamaEmbeddings

An abstract class that provides methods for embedding documents and queries using LangChain.

Hierarchy

Constructors

constructor

Properties

baseUrl caller model requestOptions?

Methods

embedDocuments embedQuery

Constructors

constructor

Properties

baseUrl

baseUrl: string = "http://localhost:11434"

caller

caller: AsyncCaller

The async caller should be used by subclasses to make any async calls, which will thus benefit from the concurrency and retry logic.

model

model: string = "llama2"

Optional requestOptions

requestOptions?: {
    embedding_only?: boolean;
    f16_kv?: boolean;
    frequency_penalty?: number;
    logits_all?: boolean;
    low_vram?: boolean;
    main_gpu?: number;
    mirostat?: number;
    mirostat_eta?: number;
    mirostat_tau?: number;
    num_batch?: number;
    num_ctx?: number;
    num_gpu?: number;
    num_gqa?: number;
    num_keep?: number;
    num_predict?: number;
    num_thread?: number;
    penalize_newline?: boolean;
    presence_penalty?: number;
    repeat_last_n?: number;
    repeat_penalty?: number;
    rope_frequency_base?: number;
    rope_frequency_scale?: number;
    stop?: string[];
    temperature?: number;
    tfs_z?: number;
    top_k?: number;
    top_p?: number;
    typical_p?: number;
    use_mlock?: boolean;
    use_mmap?: boolean;
    vocab_only?: boolean;
}

Type declaration

  • Optional embedding_only?: boolean
  • Optional f16_kv?: boolean
  • Optional frequency_penalty?: number
  • Optional logits_all?: boolean
  • Optional low_vram?: boolean
  • Optional main_gpu?: number
  • Optional mirostat?: number
  • Optional mirostat_eta?: number
  • Optional mirostat_tau?: number
  • Optional num_batch?: number
  • Optional num_ctx?: number
  • Optional num_gpu?: number
  • Optional num_gqa?: number
  • Optional num_keep?: number
  • Optional num_predict?: number
  • Optional num_thread?: number
  • Optional penalize_newline?: boolean
  • Optional presence_penalty?: number
  • Optional repeat_last_n?: number
  • Optional repeat_penalty?: number
  • Optional rope_frequency_base?: number
  • Optional rope_frequency_scale?: number
  • Optional stop?: string[]
  • Optional temperature?: number
  • Optional tfs_z?: number
  • Optional top_k?: number
  • Optional top_p?: number
  • Optional typical_p?: number
  • Optional use_mlock?: boolean
  • Optional use_mmap?: boolean
  • Optional vocab_only?: boolean

Methods

embedDocuments

  • An abstract method that takes an array of documents as input and returns a promise that resolves to an array of vectors for each document.

    Parameters

    • documents: string[]

      An array of documents to be embedded.

    Returns Promise<number[][]>

    A promise that resolves to an array of vectors for each document.

embedQuery

  • An abstract method that takes a single document as input and returns a promise that resolves to a vector for the query document.

    Parameters

    • document: string

      A single document to be embedded.

    Returns Promise<number[]>

    A promise that resolves to a vector for the query document.

Settings

Member Visibility

Theme

On This Page

Generated using TypeDoc