prompt
: Evaluation prompt used to generate the gradechoices
: List of choices/grades to choose fromchoices_scores
: Scores associated with each choiceeval_type
: One of [“classify”, “cot_classify”], determines if chain-of-thought prompting is to be applied or notprompt_var_to_column_mapping (optional)
: mapping between variables defined in the prompt vs column names in the data
How to use it?
By Default we are using GPT 3.5 Turbo. If you want to use some other model check out this tutorial