openaiRtools (version 0.2.2)

list_fine_tuning_checkpoints: List Fine-tuning Checkpoints (Convenience Function)

Description

Shortcut that creates an OpenAI client from the OPENAI_API_KEY environment variable and lists checkpoints for a fine-tuning job. Checkpoints are saved after each epoch and can be used as models directly.

Usage

list_fine_tuning_checkpoints(fine_tuning_job_id, ...)

Value

A list with $data — a list of checkpoint objects, each containing $fine_tuned_model_checkpoint (usable model ID), $step_number, and $metrics (list with $train_loss, $valid_loss, etc.).

Arguments

fine_tuning_job_id

Character. Required. The fine-tuning job ID.

...

Additional parameters passed to FineTuningCheckpointsClient$list(), such as limit and after.

Examples

if (FALSE) {
Sys.setenv(OPENAI_API_KEY = "sk-xxxxxx")

cps <- list_fine_tuning_checkpoints("ftjob-abc123")
for (cp in cps$data) {
  cat("Step:", cp$step_number, "Loss:", cp$metrics$train_loss, "\n")
}
}