做游戏代练的网站百度搜索引擎盘搜搜

错误1:openai.OpenAIError: The api_key client option must be set either by passing api_key.....

在通过openai创建客户端必须要设置api key，如果你事先已经在本机的环境中设置未起效可以手动设置，注意手动设置时不要用下面的形式

import openai
from openai import OpenAIclient = OpenAI()
openai.api_key = "YOUR API KEY"

如果你用的是client（OpenAI），这种方式不正确，正确的如下：

from openai import OpenAIclient = OpenAI(api_key="YOUR API KEY")

错误2: openai.BadRequestError: Error code: 400 - {'error': {'message':...

这个错误100%是因为你传入数据格式的原因，在这里需要参考文档来做，比如我实现的function calling，在这里需要描述函数，之前的描述为：

# 定义函数描述信息function_description = {"type": "function","function": "get_current_weather","description": "获取给定地点的当前天气","parameters": {"type": "object","properties": {"location": {"type": "string","description": "城市名称，例如：“旧金山，CA”"},"unit": {"type": "string","enum": ["celsius", "fahrenheit"]}},"required": ["location"]}}

这时候无论怎么运行都会出错，即BadRequestError，但是其实是格式出错，如果按照下面代码：

 # 定义函数描述信息function_description = [{"type": "function","function": {"name": "get_current_weather","description": "获取给定地点的当前天气","parameters": {"type": "object","properties": {"location": {"type": "string","description": "城市名称，例如：“旧金山，CA”"},"unit": {"type": "string","enum": ["celsius", "fahrenheit"]}},"required": ["location"]}}}]

即正确，完整代码如下：

import json
from openai import OpenAIdef get_current_weather(location, unit="fahrenheit"):# 实际为获取天气的接口return json.dumps({"location": location,"weather": "rain","unit": unit,})if __name__ == '__main__':# 定义客户端client = OpenAI(api_key="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")# 定义函数描述信息function_description = [{"type": "function","function": {"name": "get_current_weather","description": "获取给定地点的当前天气","parameters": {"type": "object","properties": {"location": {"type": "string","description": "城市名称，例如：“旧金山，CA”"},"unit": {"type": "string","enum": ["celsius", "fahrenheit"]}},"required": ["location"]}}}]# 发起调用response = client.chat.completions.create(model="gpt-3.5-turbo",messages=[{"role": "user", "content": "旧金山当前的天气如何？"}],tools=[function_description],temperature=0)tool_calls = response.choices[0].message.tool_calls# 如果模型确定要调用一个函数if tool_calls:# 获取模型生成的参数arguments = json.loads(tool_calls[0].function.arguments)# 调用本地函数weather_info = get_current_weather(**arguments)print(weather_info)  # 我们可以在这里看到函数调用查询到的天气信息

输出：

{"location": "San Francisco, CA", "weather": "rain", "unit": "fahrenheit"}