LASER Language-Agnostic SEntence Representations Embeddings by Meta AI

LASER is a Python library developed by the Meta AI Research team and used for creating multilingual sentence embeddings for over 147 languages as of 2/25/2024

• List of supported languages at github.com/facebookresearch/flores/blob/main/flores200/README.md#languages-in-flores-200

Dependencies

To use LaserEmbed with LangChain, install the laser_encoders Python package.

%pip install laser_encoders

Imports

from langchain_community.embeddings.laser import LaserEmbeddings

Instantiating Laser

Parameters

• lang: Optional[str]

  If empty will default to using a multilingual LASER encoder model (called “laser2”). You can find the list of supported languages and lang_codes here and here .

# Ex Instantiationz
embeddings = LaserEmbeddings(lang="eng_Latn")

Usage

Generating document embeddings

document_embeddings = embeddings.embed_documents(
    ["This is a sentence", "This is some other sentence"]
)

Generating query embeddings

query_embeddings = embeddings.embed_query("This is a query")