Feat/integration-OpenAI-autolog

[KashiwaByte]

Description

openai的api key设置需要用环境变量，临时设置：

export OPENAI_API_KEY='你的API Key'

openai<=0.28

添加了与OpenAI 0.28及以前版本的autologging集成，需要安装openai==0.28
测试代码为：

import os
import sys
import openai
from swanlab.integration.openai import autolog

autolog(init=dict(experiment_name="openai_logging123"))

chat_request_kwargs = dict(
    model="gpt-3.5-turbo",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Who won the world series in 2020?"},
        {"role": "assistant", "content": "The Los Angeles Dodgers"},
        {"role": "user", "content": "Where was it played?"},
    ],
)
response = openai.ChatCompletion.create(**chat_request_kwargs)

效果：

openai>=1.0

增加了OpenAI API的多版本集成支持，增加了Autologging对于Client类型方法改写的支持

from swanlab.integration.openai import autolog


autolog(init=dict(experiment_name="openai_autologging"))
client = autolog.client

# chat_completion

response = client.chat.completions.create(
    model="gpt-3.5-turbo-0125",
    messages=[
        {"role": "system", "content": "You are a helpful assistant designed to output JSON."},
        {"role": "user", "content": "Who won the world series in 2015?"},
    ],
)


response2 = client.chat.completions.create(
    model="gpt-3.5-turbo-0125",
    messages=[
        {"role": "system", "content": "You are a helpful assistant"},
        {"role": "user", "content": "Please write short story about ice cream?"},
    ],
)


# text_completion

response3 = client.completions.create(model="gpt-3.5-turbo-instruct", prompt="Write a song for jesus.")

Closes: #494, #500

[Zeyi-Lin]

解决差不多了