fuzzylink (version 0.2.1)

get_embeddings: Get pretrained text embeddings

Description

Get pretrained text embeddings from the OpenAI or Mistral API. Automatically batches requests to handle rate limits.

Usage

get_embeddings(
  text,
  model = "text-embedding-3-large",
  dimensions = 256,
  openai_api_key = Sys.getenv("OPENAI_API_KEY"),
  parallel = TRUE
)

Value

A matrix of embedding vectors (one per row).

Arguments

text

A character vector

model

Which embedding model to use. Defaults to 'text-embedding-3-large'.

dimensions

The dimension of the embedding vectors to return. Defaults to 256. Note that the 'mistral-embed' model will always return 1024 vectors.

openai_api_key

Your OpenAI API key. By default, looks for a system environment variable called "OPENAI_API_KEY".

parallel

TRUE to submit API requests in parallel. Setting to FALSE can reduce rate limit errors at the expense of longer runtime.

Examples

if (FALSE) {
embeddings <- get_embeddings(c('dog', 'cat', 'canine', 'feline'))
embeddings['dog',] |> dot(embeddings['canine',])
embeddings['dog',] |> dot(embeddings['feline',])
}