generate_ratings: Generate Ratings for a Stimulus Using LLM APIs

In chatRater: Rating and Evaluating Texts Using Large Language Models

Generate Ratings for a Stimulus Using LLM APIs

Description

Generates ratings for a given stimulus by calling one of several LLM APIs.

Usage

generate_ratings(
  model = "gpt-3.5-turbo",
  stim = "kick the bucket",
  prompt = "You are a native English speaker.",
  question = "Please rate the following stim:",
  top_p = 1,
  temp = 0,
  n_iterations = 30,
  api_key = "",
  debug = FALSE
)

Arguments

model	A character string specifying the LLM model (e.g., "gpt-3.5-turbo", "deepseek-chat").
stim	A character string representing the stimulus (e.g., an idiom).
prompt	A character string for the system prompt (e.g., "You are a native English speaker.").
question	A character string for the user prompt (e.g., "Please rate the following stim:").
top_p	Numeric value for the probability mass (default 1).
temp	Numeric value for the temperature (default 0).
n_iterations	An integer indicating how many times to query the API.
api_key	API key as a character string.
debug	Logical; if TRUE, debug information is printed.

Details

This function supports multiple APIs. Branching is based on the model parameter.

Value

A data frame containing the stimulus, the rating, and the iteration number.

Examples

## Not run: 
  ratings <- generate_ratings(
    model = "gpt-3.5-turbo",
    stim = "kick the bucket",
    prompt = "You are a native English speaker.",
    question = "Please rate the following stim:",
    n_iterations = 5,
    api_key = "your_api_key"
  )
  print(ratings)

## End(Not run)

chatRater documentation built on April 4, 2025, 1:03 a.m.