rakam_systems.vector_store.VectorStores

class rakam_systems.vector_store.VectorStores(base_index_path: str, embedding_model: str)

A class for managing multiple vector stores using FAISS and SentenceTransformers.

__init__(base_index_path: str, embedding_model: str) → None

Initializes the VectorStores with the specified base index path and embedding model.

Parameters:

• base_index_path – Base path to store the FAISS indexes.

• embedding_model – Pre-trained SentenceTransformer model name.

Methods

__init__(base_index_path, embedding_model)	Initializes the VectorStores with the specified base index path and embedding model.
create_from_files(stores_files)	Creates FAISS indexes from dictionaries of store names and VSFile objects.
create_from_nodes(store_name, nodes)	Creates a FAISS index from a list of nodes and stores it under the given store name.
get_embeddings(sentences[, parallel])	Generates embeddings for a list of sentences.
load_all_stores()	Loads all vector stores from the base directory.
load_store(store_path)	Loads a single vector store from the specified directory.
predict_embeddings(query)	Predicts embeddings for a given query using the embedding model.
search(store_name, query[, distance_type, ...])	Searches the specified store for the closest embeddings to the query.