つくよみちゃんの会話テキストデータセットでGPT-3のファインチューニングを試す

import pandas as pd

# データセットの準備
df = pd.read_csv(
    'tsukuyomi.csv', 
    usecols=[1,2], 
    names=['prompt','completion'], 
    skiprows=2)
df.head()

# JSONLファイルに出力
df.to_json("tsukuyomi.jsonl", orient='records', lines=True)

# openaiパッケージのインストール
!pip install --upgrade openai

# APIキーの設定
%env OPENAI_API_KEY=<API_KEY>

# データセットの検証
!openai tools fine_tunes.prepare_data -f tsukuyomi.jsonl -q

Analyzing...

- Your file contains 469 prompt-completion pairs
- There are 2 duplicated prompt-completion sets. These are rows: [438, 439]
- More than a third of your `prompt` column/key is uppercase. Uppercase prompts tends to perform worse than a mixture of case encountered in normal language. We recommend to lower case the data if that makes sense in your domain. See https://beta.openai.com/docs/guides/fine-tuning/preparing-your-dataset for more details
- More than a third of your `completion` column/key is uppercase. Uppercase completions tends to perform worse than a mixture of case encountered in normal language. We recommend to lower case the data if that makes sense in your domain. See https://beta.openai.com/docs/guides/fine-tuning/preparing-your-dataset for more details
- 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
- Your data does not contain a common ending at the end of your completions. Having a common ending string appended to the end of the completion makes it clearer to the fine-tuned model where the completion should end. See https://beta.openai.com/docs/guides/fine-tuning/preparing-your-dataset for more detail and examples.
- 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] Remove 2 duplicate rows [Y/n]: Y
- [Recommended] Lowercase all your data in column/key `prompt` [Y/n]: Y
/usr/local/lib/python3.7/dist-packages/openai/validators.py:448: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead


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

Wrote modified file to `tsukuyomi_prepared.jsonl`
Feel free to take a look!

Now use that file when fine-tuning:
> openai api fine_tunes.create -t "tsukuyomi_prepared.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 26.16 minutes to train a `curie` model, and less for `ada` and `babbage`. Queue will approximately take half an hour per job ahead of you.

# ファインチューニングの実行
!openai api fine_tunes.create \
    -t "tsukuyomi_prepared.jsonl" \
    -m davinci

[2022-11-09 13:29:47] Fine-tune costs $3.21
Job complete! Status: succeeded 🎉
Try out your fine-tuned model:

openai api completions.create -m davinci:ft-personal-2022-11-09-13-41-19 -p <YOUR_PROMPT>

import openai

prompt='今日はいい天気だね->'
response = openai.Completion.create(
    engine='davinci:ft-personal-2022-11-09-13-41-19', 
    prompt=prompt, 
    max_tokens=100,
    stop='\n')
print(prompt+response['choices'][0]['text'])

今日はいい天気だね->そうですね！　暖かくてよいですね！