Struct OpenAIEmbeddingClient

pub struct OpenAIEmbeddingClient { /* private fields */ }

OpenAIEmbeddingClient

Allows for interacting with the OpenAI API to generate embeddings. You can either embed a single string or a batch of strings.

Examples

use rag_toolchain::common::*;
use rag_toolchain::clients::*;
async fn generate_embedding() {
    let client: OpenAIEmbeddingClient = OpenAIEmbeddingClient::try_new(OpenAIEmbeddingModel::TextEmbeddingAda002).unwrap();
    let chunk: Chunk = Chunk::new("this would be the text you are embedding");
    let embedding: Embedding = client.generate_embedding(chunk).await.unwrap();
    // This would be the vector representation of the text
    let vector: Vec<f32> = embedding.vector();
}

Required Environment Variables

OPENAI_API_KEY: The API key to use for the OpenAI API

Implementations

impl OpenAIEmbeddingClient

pub fn try_new( embedding_model: OpenAIEmbeddingModel, ) -> Result<OpenAIEmbeddingClient, VarError>

OpenAIEmbeddingClient::try_new

Constructor to create a new OpenAIEmbeddingClient. This will fail if the OPENAI_API_KEY environment variable is not set.

Arguments

• embedding_model: OpenAIEmbeddingModel - The model to use for the embeddings

Errors

• VarError - If the OPENAI_API_KEY environment variable is not set.

Returns

• OpenAIEmbeddingClient - The newly created OpenAIEmbeddingClient

Trait Implementations

impl AsyncEmbeddingClient for OpenAIEmbeddingClient

async fn generate_embeddings( &self, text: Chunks, ) -> Result<Vec<Embedding>, OpenAIError>

OpenAIEmbeddingClient::generate_embeddings

Function to generate embeddings for Chunks. Allows you to get an embedding for multiple strings.

Arguments

• text: Chunk - The text chunks/strings to generate an embeddings for.

Errors

• OpenAIError - If the request to OpenAI fails.

Returns

• Vec<Embedding> - A result containing pairs of the original text and the embedding that was generated.

async fn generate_embedding( &self, text: Chunk, ) -> Result<Embedding, Self::ErrorType>

OpenAIEmbeddingClient::generate_embedding

Function to generate an embedding for a Chunk. Allows you to get an embedding for a single string.

Arguments

• text: Chunk - The text chunk/string to generate an embedding for.

Errors

• OpenAIError - If the request to OpenAI fails.

Returns

• Embedding - the generated embedding

type ErrorType = OpenAIError