Struct EmbeddingsBuilder

pub struct EmbeddingsBuilder<M: EmbeddingModel> { /* private fields */ }

Builder for creating a collection of embeddings

Implementations

impl<M: EmbeddingModel> EmbeddingsBuilder<M>

pub fn new(model: M) -> Self

Create a new embedding builder with the given embedding model

pub fn simple_document(self, id: &str, document: &str) -> Self

Add a simple document to the embedding collection. The provided document string will be used for the embedding.

pub fn simple_documents(self, documents: Vec<(String, String)>) -> Self

Add multiple documents to the embedding collection. Each element of the vector is a tuple of the form (id, document).

pub fn tool( self, tool: impl ToolEmbedding + 'static, ) -> Result<Self, EmbeddingError>

Add a tool to the embedding collection. The tool.context() corresponds to the document being stored while tool.embedding_docs() corresponds to the documents that will be used to generate the embeddings.

pub fn tools(self, toolset: &ToolSet) -> Result<Self, EmbeddingError>

Add the tools from the given toolset to the embedding collection.

pub fn document<T: Serialize>( self, id: &str, document: T, embed_documents: Vec<String>, ) -> Self

Add a document to the embedding collection. embed_documents are the documents that will be used to generate the embeddings for document.

pub fn documents<T: Serialize>( self, documents: Vec<(String, T, Vec<String>)>, ) -> Self

Add multiple documents to the embedding collection. Each element of the vector is a tuple of the form (id, document, embed_documents).

pub fn json_document( self, id: &str, document: Value, embed_documents: Vec<String>, ) -> Self

Add a json document to the embedding collection.

pub fn json_documents( self, documents: Vec<(String, Value, Vec<String>)>, ) -> Self

Add multiple json documents to the embedding collection.

pub async fn build(self) -> Result<Vec<DocumentEmbeddings>, EmbeddingError>

Generate the embeddings for the given documents