antimatter.handlers.langchain

Module Contents

Classes

EmbeddingClient
BaseDocument	BaseDocument is a wrapper around the langchain Document.
LangchainHandler	The LangchainHandler DataHandler supports an implementation of a langchain

class antimatter.handlers.langchain.EmbeddingClient

OPEN_AI = 'openai'

HUGGING_FACE = 'huggingface'

class antimatter.handlers.langchain.BaseDocument

BaseDocument is a wrapper around the langchain Document. It is used to store the page content of the document.

content: Any

embedding: List[float]

class antimatter.handlers.langchain.LangchainHandler

Bases: antimatter.handlers.DataHandler

The LangchainHandler DataHandler supports an implementation of a langchain Retriever. This handler assumes that the underlying data is a list of two-dimensional data.

from_generic(cols: List[str], generic_data: List[List[bytes]], extra: Dict[str, Any]) → Any

from_generic loads the generic data into an implementation of a langchain Retriever with langchain Embeddings.

Parameters:

• cols – the column names for the underlying data

• generic_data – the capsule’s generic data format that is loaded into a langchain Retriever

• extra – extra data for the DataHandler, containing information for the Embeddings

Returns:

the langchain Retriever built with the dataset

to_generic(data: Any) → Tuple[List[str], List[List[bytes]], Dict[str, Any]]

to_generic converts a langchain Retriever with langchain Embeddings into the generic data format.

Parameters:

data – the langchain Retriever

Returns:

the data in its generic format