Skip to content

vectors #

Classes#

VectorisedGraph#

VectorisedGraph object that contains embedded documents that correspond to graph entities.

Document#

A document corresponding to a graph entity. Used to generate embeddings.

Embedding#

VectorSelection#

OpenAIEmbeddings#

OpenAI-compatible embedding configuration. Pass an instance of this to VectorCache(...) to drive vectorise(...).

VectorCache#

Cache wrapping an embedding model. Pass to Graph.vectorise(model=...) or other vectorisation entry points.

EmbeddingServer#

RunningEmbeddingServer#

Functions#

embedding_server(function) #

Wrap a Python callable so it can be served as an OpenAI-compatible embedding endpoint via EmbeddingServer.serve(...).

Parameters:

Name Type Description Default
function Callable[[str], list[float]]

A callable that maps a text input to its embedding vector.

required

Returns:

Type Description
EmbeddingServer