langchain.vectorstores.marqo.Marqo

class langchain.vectorstores.marqo.Marqo(client: marqo.Client, index_name: str, add_documents_settings: Optional[Dict[str, Any]] = None, searchable_attributes: Optional[List[str]] = None, page_content_builder: Optional[Callable[[Dict[str, Any]], str]] = None)[source]

Marqo vector store.

Marqo indexes have their own models associated with them to generate your embeddings. This means that you can selected from a range of different models and also use CLIP models to create multimodal indexes with images and text together.

Marqo also supports more advanced queries with multiple weighted terms, see See https://docs.marqo.ai/latest/#searching-using-weights-in-queries. This class can flexibly take strings or dictionaries for weighted queries in its similarity search methods.

To use, you should have the marqo python package installed, you can do this with pip install marqo.

Example

import marqo
from langchain.vectorstores import Marqo
client = marqo.Client(url=os.environ["MARQO_URL"], ...)
vectorstore = Marqo(client, index_name)

Initialize with Marqo client.

Attributes

embeddings

Access the query embedding object if available.

Methods

__init__(client, index_name[, ...])

Initialize with Marqo client.

aadd_documents(documents, **kwargs)

Run more documents through the embeddings and add to the vectorstore.

aadd_texts(texts[, metadatas])

Run more texts through the embeddings and add to the vectorstore.

add_documents(documents, **kwargs)

Run more documents through the embeddings and add to the vectorstore.

add_texts(texts[, metadatas])

Upload texts with metadata (properties) to Marqo.

adelete([ids])

Delete by vector ID or other criteria.

afrom_documents(documents, embedding, **kwargs)

Return VectorStore initialized from documents and embeddings.

afrom_texts(texts, embedding[, metadatas])

Return VectorStore initialized from texts and embeddings.

amax_marginal_relevance_search(query[, k, ...])

Return docs selected using the maximal marginal relevance.

amax_marginal_relevance_search_by_vector(...)

Return docs selected using the maximal marginal relevance.

as_retriever(**kwargs)

Return VectorStoreRetriever initialized from this VectorStore.

asearch(query, search_type, **kwargs)

Return docs most similar to query using specified search type.

asimilarity_search(query[, k])

Return docs most similar to query.

asimilarity_search_by_vector(embedding[, k])

Return docs most similar to embedding vector.

asimilarity_search_with_relevance_scores(query)

Return docs and relevance scores in the range [0, 1], asynchronously.

asimilarity_search_with_score(*args, **kwargs)

Run similarity search with distance asynchronously.

bulk_similarity_search(queries[, k])

Search the marqo index for the most similar documents in bulk with multiple queries.

bulk_similarity_search_with_score(queries[, k])

Return documents from Marqo that are similar to the query as well as their scores using a batch of queries.

delete([ids])

Delete by vector ID or other criteria.

from_documents(documents[, embedding])

Return VectorStore initialized from documents.

from_texts(texts[, embedding, metadatas, ...])

Return Marqo initialized from texts.

get_indexes()

Helper to see your available indexes in marqo, useful if the from_texts method was used without an index name specified

get_number_of_documents()

Helper to see the number of documents in the index

marqo_bulk_similarity_search(queries[, k])

Return documents from Marqo using a bulk search, exposes Marqo's output directly

marqo_similarity_search(query[, k])

Return documents from Marqo exposing Marqo's output directly

max_marginal_relevance_search(query[, k, ...])

Return docs selected using the maximal marginal relevance.

max_marginal_relevance_search_by_vector(...)

Return docs selected using the maximal marginal relevance.

search(query, search_type, **kwargs)

Return docs most similar to query using specified search type.

similarity_search(query[, k])

Search the marqo index for the most similar documents.

similarity_search_by_vector(embedding[, k])

Return docs most similar to embedding vector.

similarity_search_with_relevance_scores(query)

Return docs and relevance scores in the range [0, 1].

similarity_search_with_score(query[, k])

Return documents from Marqo that are similar to the query as well as their scores.

__init__(client: marqo.Client, index_name: str, add_documents_settings: Optional[Dict[str, Any]] = None, searchable_attributes: Optional[List[str]] = None, page_content_builder: Optional[Callable[[Dict[str, Any]], str]] = None)[source]

Initialize with Marqo client.

async aadd_documents(documents: List[Document], **kwargs: Any) List[str]

Run more documents through the embeddings and add to the vectorstore.

Parameters

(List[Document] (documents) – Documents to add to the vectorstore.

Returns

List of IDs of the added texts.

Return type

List[str]

async aadd_texts(texts: Iterable[str], metadatas: Optional[List[dict]] = None, **kwargs: Any) List[str]

Run more texts through the embeddings and add to the vectorstore.

add_documents(documents: List[Document], **kwargs: Any) List[str]

Run more documents through the embeddings and add to the vectorstore.

Parameters

(List[Document] (documents) – Documents to add to the vectorstore.

Returns

List of IDs of the added texts.

Return type

List[str]

add_texts(texts: Iterable[str], metadatas: Optional[List[dict]] = None, **kwargs: Any) List[str][source]

Upload texts with metadata (properties) to Marqo.

You can either have marqo generate ids for each document or you can provide your own by including a “_id” field in the metadata objects.

Parameters
  • texts (Iterable[str]) – am iterator of texts - assumed to preserve an

  • metadatas. (order that matches the) –

  • metadatas (Optional[List[dict]], optional) – a list of metadatas.

Raises
  • ValueError – if metadatas is provided and the number of metadatas differs

  • from the number of texts.

Returns

The list of ids that were added.

Return type

List[str]

async adelete(ids: Optional[List[str]] = None, **kwargs: Any) Optional[bool]

Delete by vector ID or other criteria.

Parameters
  • ids – List of ids to delete.

  • **kwargs – Other keyword arguments that subclasses might use.

Returns

True if deletion is successful, False otherwise, None if not implemented.

Return type

Optional[bool]

async classmethod afrom_documents(documents: List[Document], embedding: Embeddings, **kwargs: Any) VST

Return VectorStore initialized from documents and embeddings.

async classmethod afrom_texts(texts: List[str], embedding: Embeddings, metadatas: Optional[List[dict]] = None, **kwargs: Any) VST

Return VectorStore initialized from texts and embeddings.

Return docs selected using the maximal marginal relevance.

async amax_marginal_relevance_search_by_vector(embedding: List[float], k: int = 4, fetch_k: int = 20, lambda_mult: float = 0.5, **kwargs: Any) List[Document]

Return docs selected using the maximal marginal relevance.

as_retriever(**kwargs: Any) VectorStoreRetriever

Return VectorStoreRetriever initialized from this VectorStore.

Parameters
  • search_type (Optional[str]) – Defines the type of search that the Retriever should perform. Can be “similarity” (default), “mmr”, or “similarity_score_threshold”.

  • search_kwargs (Optional[Dict]) –

    Keyword arguments to pass to the search function. Can include things like:

    k: Amount of documents to return (Default: 4) score_threshold: Minimum relevance threshold

    for similarity_score_threshold

    fetch_k: Amount of documents to pass to MMR algorithm (Default: 20) lambda_mult: Diversity of results returned by MMR;

    1 for minimum diversity and 0 for maximum. (Default: 0.5)

    filter: Filter by document metadata

Returns

Retriever class for VectorStore.

Return type

VectorStoreRetriever

Examples:

# Retrieve more documents with higher diversity
# Useful if your dataset has many similar documents
docsearch.as_retriever(
    search_type="mmr",
    search_kwargs={'k': 6, 'lambda_mult': 0.25}
)

# Fetch more documents for the MMR algorithm to consider
# But only return the top 5
docsearch.as_retriever(
    search_type="mmr",
    search_kwargs={'k': 5, 'fetch_k': 50}
)

# Only retrieve documents that have a relevance score
# Above a certain threshold
docsearch.as_retriever(
    search_type="similarity_score_threshold",
    search_kwargs={'score_threshold': 0.8}
)

# Only get the single most similar document from the dataset
docsearch.as_retriever(search_kwargs={'k': 1})

# Use a filter to only retrieve documents from a specific paper
docsearch.as_retriever(
    search_kwargs={'filter': {'paper_title':'GPT-4 Technical Report'}}
)
async asearch(query: str, search_type: str, **kwargs: Any) List[Document]

Return docs most similar to query using specified search type.

Return docs most similar to query.

async asimilarity_search_by_vector(embedding: List[float], k: int = 4, **kwargs: Any) List[Document]

Return docs most similar to embedding vector.

async asimilarity_search_with_relevance_scores(query: str, k: int = 4, **kwargs: Any) List[Tuple[Document, float]]

Return docs and relevance scores in the range [0, 1], asynchronously.

0 is dissimilar, 1 is most similar.

Parameters
  • query – input text

  • k – Number of Documents to return. Defaults to 4.

  • **kwargs

    kwargs to be passed to similarity search. Should include: score_threshold: Optional, a floating point value between 0 to 1 to

    filter the resulting set of retrieved docs

Returns

List of Tuples of (doc, similarity_score)

async asimilarity_search_with_score(*args: Any, **kwargs: Any) List[Tuple[Document, float]]

Run similarity search with distance asynchronously.

Search the marqo index for the most similar documents in bulk with multiple queries.

Parameters
  • queries (Iterable[Union[str, Dict[str, float]]]) – An iterable of queries to

  • bulk (execute in) –

  • of (queries in the list can be strings or dictionaries) –

  • queries. (weighted) –

  • k (int, optional) – The number of documents to return for each query.

  • 4. (Defaults to) –

Returns

A list of results for each query.

Return type

List[List[Document]]

bulk_similarity_search_with_score(queries: Iterable[Union[str, Dict[str, float]]], k: int = 4, **kwargs: Any) List[List[Tuple[Document, float]]][source]

Return documents from Marqo that are similar to the query as well as their scores using a batch of queries.

Parameters
  • query (Iterable[Union[str, Dict[str, float]]]) – An iterable of queries

  • bulk (to execute in) –

  • dictionaries (queries in the list can be strings or) –

  • queries. (of weighted) –

  • k (int, optional) – The number of documents to return. Defaults to 4.

Returns

A list of lists of the matching documents and their scores for each query

Return type

List[Tuple[Document, float]]

delete(ids: Optional[List[str]] = None, **kwargs: Any) Optional[bool]

Delete by vector ID or other criteria.

Parameters
  • ids – List of ids to delete.

  • **kwargs – Other keyword arguments that subclasses might use.

Returns

True if deletion is successful, False otherwise, None if not implemented.

Return type

Optional[bool]

classmethod from_documents(documents: List[Document], embedding: Optional[Embeddings] = None, **kwargs: Any) Marqo[source]

Return VectorStore initialized from documents. Note that Marqo does not need embeddings, we retain the parameter to adhere to the Liskov substitution principle.

Parameters
  • documents (List[Document]) – Input documents

  • embedding (Any, optional) – Embeddings (not required). Defaults to None.

Returns

A Marqo vectorstore

Return type

VectorStore

classmethod from_texts(texts: List[str], embedding: Any = None, metadatas: Optional[List[dict]] = None, index_name: str = '', url: str = 'http://localhost:8882', api_key: str = '', add_documents_settings: Optional[Dict[str, Any]] = None, searchable_attributes: Optional[List[str]] = None, page_content_builder: Optional[Callable[[Dict[str, str]], str]] = None, index_settings: Optional[Dict[str, Any]] = None, verbose: bool = True, **kwargs: Any) Marqo[source]

Return Marqo initialized from texts. Note that Marqo does not need embeddings, we retain the parameter to adhere to the Liskov substitution principle.

This is a quick way to get started with marqo - simply provide your texts and metadatas and this will create an instance of the data store and index the provided data.

To know the ids of your documents with this approach you will need to include them in under the key “_id” in your metadatas for each text

Example: .. code-block:: python

from langchain.vectorstores import Marqo

datastore = Marqo(texts=[‘text’], index_name=’my-first-index’, url=’http://localhost:8882’)

Parameters
  • texts (List[str]) – A list of texts to index into marqo upon creation.

  • embedding (Any, optional) – Embeddings (not required). Defaults to None.

  • index_name (str, optional) – The name of the index to use, if none is

  • None. (accompany the texts. Defaults to) –

  • url (str, optional) – The URL for Marqo. Defaults to “http://localhost:8882”.

  • api_key (str, optional) – The API key for Marqo. Defaults to “”.

  • metadatas (Optional[List[dict]], optional) – A list of metadatas, to

  • None.

  • Can (this is only used when a new index is being created. Defaults to "cpu".) –

  • "cuda". (be "cpu" or) –

  • add_documents_settings (Optional[Dict[str, Any]], optional) – Settings

  • documents (for adding) –

  • see

  • https – //docs.marqo.ai/0.0.16/API-Reference/documents/#query-parameters.

  • {}. (Defaults to) –

  • index_settings (Optional[Dict[str, Any]], optional) – Index settings if

  • exist (the index doesn't) –

  • see

  • https – //docs.marqo.ai/0.0.16/API-Reference/indexes/#index-defaults-object.

  • {}.

Returns

An instance of the Marqo vector store

Return type

Marqo

get_indexes() List[Dict[str, str]][source]

Helper to see your available indexes in marqo, useful if the from_texts method was used without an index name specified

Returns

The list of indexes

Return type

List[Dict[str, str]]

get_number_of_documents() int[source]

Helper to see the number of documents in the index

Returns

The number of documents

Return type

int

Return documents from Marqo using a bulk search, exposes Marqo’s output directly

Parameters
  • queries (Iterable[Union[str, Dict[str, float]]]) – A list of queries.

  • k (int, optional) – The number of documents to return for each query.

  • 4. (Defaults to) –

Returns

A bulk search results object

Return type

Dict[str, Dict[List[Dict[str, Dict[str, Any]]]]]

Return documents from Marqo exposing Marqo’s output directly

Parameters
  • query (str) – The query to search with.

  • k (int, optional) – The number of documents to return. Defaults to 4.

Returns

This hits from marqo.

Return type

List[Dict[str, Any]]

Return docs selected using the maximal marginal relevance.

Maximal marginal relevance optimizes for similarity to query AND diversity among selected documents.

Parameters
  • query – Text to look up documents similar to.

  • k – Number of Documents to return. Defaults to 4.

  • fetch_k – Number of Documents to fetch to pass to MMR algorithm.

  • lambda_mult – Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5.

Returns

List of Documents selected by maximal marginal relevance.

max_marginal_relevance_search_by_vector(embedding: List[float], k: int = 4, fetch_k: int = 20, lambda_mult: float = 0.5, **kwargs: Any) List[Document]

Return docs selected using the maximal marginal relevance.

Maximal marginal relevance optimizes for similarity to query AND diversity among selected documents.

Parameters
  • embedding – Embedding to look up documents similar to.

  • k – Number of Documents to return. Defaults to 4.

  • fetch_k – Number of Documents to fetch to pass to MMR algorithm.

  • lambda_mult – Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5.

Returns

List of Documents selected by maximal marginal relevance.

search(query: str, search_type: str, **kwargs: Any) List[Document]

Return docs most similar to query using specified search type.

Search the marqo index for the most similar documents.

Parameters
  • query (Union[str, Dict[str, float]]) – The query for the search, either

  • query. (as a string or a weighted) –

  • k (int, optional) – The number of documents to return. Defaults to 4.

Returns

k documents ordered from best to worst match.

Return type

List[Document]

similarity_search_by_vector(embedding: List[float], k: int = 4, **kwargs: Any) List[Document]

Return docs most similar to embedding vector.

Parameters
  • embedding – Embedding to look up documents similar to.

  • k – Number of Documents to return. Defaults to 4.

Returns

List of Documents most similar to the query vector.

similarity_search_with_relevance_scores(query: str, k: int = 4, **kwargs: Any) List[Tuple[Document, float]]

Return docs and relevance scores in the range [0, 1].

0 is dissimilar, 1 is most similar.

Parameters
  • query – input text

  • k – Number of Documents to return. Defaults to 4.

  • **kwargs

    kwargs to be passed to similarity search. Should include: score_threshold: Optional, a floating point value between 0 to 1 to

    filter the resulting set of retrieved docs

Returns

List of Tuples of (doc, similarity_score)

similarity_search_with_score(query: Union[str, Dict[str, float]], k: int = 4) List[Tuple[Document, float]][source]

Return documents from Marqo that are similar to the query as well as their scores.

Parameters
  • query (str) – The query to search with, either as a string or a weighted

  • query.

  • k (int, optional) – The number of documents to return. Defaults to 4.

Returns

The matching documents and their scores, ordered by descending score.

Return type

List[Tuple[Document, float]]

Examples using Marqo