max_tokens=1
stop=['tive']

import openai
import os

# OPENAI_ORG_KEY, OPENAI_API_KEY はあらかじめ環境変数に設定しておく
openai.organization = os.getenv("OPENAI_ORG_KEY")
openai.api_key = os.getenv("OPENAI_API_KEY")

def completion(x, model):
     pre_string1 = "Classify following sentence as Positive or Negative\n#####\n"
     pre_string2 = "There are films that make careers. For George Romero, it was NIGHT OF THE LIVING DEAD; for Kevin Smith, "\
     "CLERKS; for Robert Rodriguez, EL MARIACHI. Add to that list Onur Tukel's absolutely amazing DING-A-LING-LESS. Flawless "\
     "film-making, and as assured and as professional as any of the aforementioned movies. I haven't laughed this hard since "\
     "I saw THE FULL MONTY. (And, even then, I don't think I laughed quite this hard... So to speak.) Tukel's talent is "\
     "considerable: DING-A-LING-LESS is so chock full of double entendres that one would have to sit down with a copy of "\
     "this script and do a line-by-line examination of it to fully appreciate the, uh, breadth and width of it. Every shot "\
     "is beautifully composed (a clear sign of a sure-handed director), and the performances all around are solid (there's "\
     "none of the over-the-top scenery chewing one might've expected from a film like this). DING-A-LING-LESS is a film whose "\
     "time has come."
     prompt = pre_string1 + pre_string2 + " ->" + " Positive" + "\n#####\n" + x + " ->"
     max_tokens = 1
     stop = ['tive']

     response = openai.Completion.create( 
           model=model,
           prompt=prompt,
           max_tokens=max_tokens,
           stop=stop
     )

     return response['choices'][0]['text'] + "tive"

model = "curie" # "babbage", "curie", "davinci" から指定

# Completion 実行
df_test.loc[:,'inference'] = df_test.prompt.apply(completion, args=(model, ))

# 正誤判定
df_test.inference =df_test.inference.map(lambda x: x.replace(' ',''))
df_test.loc[:,'judge'] = df_test.completion == df_test.inference

print(collections.Counter(df_test.judge.values)) # Counter({True: 60, False: 40})

{"prompt": "プロンプトテキスト", "completion": "生成するテキスト"}
{"prompt": "プロンプトテキスト", "completion": "生成するテキスト"}
{"prompt": "プロンプトテキスト", "completion": "生成するテキスト"}
...

file_name = 'IMDB_train_500_for_fine_tune.jsonl'
train_jsonl = df_train[:500].to_json(orient='records', force_ascii=False, lines=True)
with open(file_name, mode='w') as f:
      f.write(train_jsonl)

$ openai tools fine_tunes.prepare_data -f IMDB_train_500_for_fine_tune.jsonl

Analyzing...

- Your file contains 500 prompt-completion pairs
- Based on your data it seems like you're trying to fine-tune a model for classification
- For classification, we recommend you try one of the faster and cheaper models, such as `ada`
- For classification, you can estimate the expected model performance by keeping a held out dataset,
which is not used for training
- Your data does not contain a common separator at the end of your prompts. Having a separator string
appended to the end of the prompt makes it clearer to the fine-tuned model where the completion should begin.
See https://beta.openai.com/docs/guides/fine-tuning/preparing-your-dataset for more detail and examples.
If you intend to do open-ended generation, then you should leave the prompts empty
- The completion should start with a whitespace character (` `). This tends to produce better results due to
the tokenization we use.
See https://beta.openai.com/docs/guides/fine-tuning/preparing-your-dataset for more details

Based on the analysis we will perform the following actions:
- [Recommended] Add a suffix separator ` ->` to all prompts [Y/n]: Y
- [Recommended] Add a whitespace character to the beginning of the completion [Y/n]: Y
- [Recommended] Would you like to split into training and validation set? [Y/n]: Y

Your data will be written to a new JSONL file. Proceed [Y/n]: Y

Wrote modified files to
`IMDB_train_500_for_fine_tune_prepared_train.jsonl` and `IMDB_train_500_for_fine_tune_prepared_valid.jsonl`
Feel free to take a look!

Now use that file when fine-tuning:
> openai api fine_tunes.create -t "IMDB_train_500_for_fine_tune_prepared_train.jsonl"
-v "IMDB_train_500_for_fine_tune_prepared_valid.jsonl" --compute_classification_metrics
--classification_positive_class " Negative"

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=["tive"]` so that the generated texts ends at the expected place.
Once your model starts training, it'll approximately take 14.33 minutes to train a `curie` model,
and less for `ada` and `babbage`. Queue will approximately take half an hour per job ahead of you.

$ openai api fine_tunes.create -t "IMDB_train_500_for_fine_tune_prepared_train.jsonl"
-v "IMDB_train_500_for_fine_tune_prepared_valid.jsonl" --compute_classification_metrics
--classification_positive_class " Positive"

def completion(x, model):
     prompt = x + " ->"
     model = model
     max_tokens = 1
     stop = ['tive']

     response = openai.Completion.create(
           model=model,
           prompt=prompt,
           max_tokens=max_tokens,
           stop=stop
     )

     return response['choices'][0]['text'] + "tive"

# "curie" 500 training data
model = "curie:ft-networks-company-macnica-inc-2022-02-04-15-53-20"

# Completion 実行
df_test.loc[:,'inference'] = df_test.prompt.apply(completion, args=(model, ))

# 正誤判定
df_test.inference = df_test.inference.map(lambda x: x.replace(' ',''))
df_test.loc[:, 'judge'] = df_test.completion == df_test.inference

print(collections.Counter(df_test.judge.values)) # Counter({True: 91, False: 9})