How to fine tune a model using terminal

Documentation
,
1

I have a file called abc.csv. I made it into a jsonl using: openai tools fine_tunes.prepare_data -f we-wingit-data.csv and I got this below:
Wrote modified file to we-wingit-data_prepared.jsonl
Feel free to take a look!

Now use that file when fine-tuning:

openai api fine_tunes.create -t “abc.jsonl”

After you’ve fine-tuned a model, remember that your prompt has to end with the indicator string -> for the model to start generating completions, rather than continuing with the prompt. Make sure to include stop=["\n"] so that the generated texts ends at the expected place.
Once your model starts training, it’ll approximately take 2.97 minutes to train a curie model, and less for ada and babbage. Queue will approximately take half an hour per job ahead of you.
Now I am trying to fine tune the jsonl file using comand :
openai api fine_tunes.create -t abc_prepared.jsonl -m gpt-3.5-turbo

but I get error:
usage: openai api [-h]
{chat.completions.create,images.generate,images.edit,images.create_variation,audio.transcriptions.create,audio.translations.create,files.create,files.retrieve,files.delete,files.list,models.list,models.retrieve,models.delete,completions.create}
…
openai api: error: argument {chat.completions.create,images.generate,images.edit,images.create_variation,audio.transcriptions.create,audio.translations.create,files.create,files.retrieve,files.delete,files.list,models.list,models.retrieve,models.delete,completions.create}: invalid choice: ‘fine_tunes.create’ (choose from ‘chat.completions.create’, ‘images.generate’, ‘images.edit’, ‘images.create_variation’, ‘audio.transcriptions.create’, ‘audio.translations.create’, ‘files.create’, ‘files.retrieve’, ‘files.delete’, ‘files.list’, ‘models.list’, ‘models.retrieve’, ‘models.delete’, ‘completions.create’)

I also tried to get the id and them use file id instead of file name, got the same issue. How to fix this?

2

Heya! Welcome to the forum.

What version of the library are you using?

pip install --upgrade openai

3

I did make an upgrade. My current version is openai 1.33.0

But still stuck on how to do it. Could not find anything in the docs either

4

Try to use the OpenAI client within a Python script?

from openai import OpenAI
client = OpenAI()

client.fine_tuning.jobs.create(
  training_file="file-abc123", 
  model="gpt-3.5-turbo"
)

You can fine more on how to fine-tune in the docs.

5

Yeah I saw this in documentation but been trying to use CLI as I feel its more neat to use