openai-api-call 0.1.0

A short wrapper of the OpenAI api call.

以下为 ChatGPT 自动生成的 README.md 文件（待修改），中文文档移步这里。

Openai API call

A simple wrapper for OpenAI API, which can send prompt message and return response.

Installation

pip install git+https://github.com/RexWzh/openai_api_call.git

Usage

Set API Key

import openai
openai.api_key = "sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

Or set OPENAI_API_KEY in ~/.bashrc to automatically set it when you start the terminal:

# Add the following code to ~/.bashrc
export OPENAI_API_KEY="sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

Set Proxy (Optional)

from openai_api_call import proxy_on, proxy_off, show_proxy
# Check the current proxy
show_proxy()

# Set local proxy, port number is 7890 by default
proxy_on("127.0.0.1", port=7890)

# Check the updated proxy
show_proxy()

# Turn off proxy
proxy_off() 

Basic Usage

Example 1, send prompt and return information:

from openai_api_call import prompt2response, show_apikey

# Check if API key is set
show_apikey()

# Check if proxy is enabled
show_proxy()

# Send prompt and return response
prompt = "Hello, GPT-3.5!"
resp = prompt2response(prompt)
print(resp.content)

Example 2, customize the message template and return the information and the number of consumed tokens:

import openai_api_call

# Customize the sending template
openai_api_call.default_prompt = lambda msg: [
    {"role": "system", "content": "帮我翻译这段文字"},
    {"role": "user", "content": msg}
]
prompt = "Hello!"
# Set the number of retries to Inf
response = prompt2response(prompt, temperature=0.5, max_requests=-1)
print("Number of consumed tokens: ", response.total_tokens)
print("Returned content: ", response.content)

Advanced Usage

Continue chatting based on the last response:

# first call
prompt = "Hello, GPT-3.5!"
resp = prompt2response(prompt)
print(resp.content)

# next call
next_prompt = resp.next_prompt("How are you?")
print(next_prompt)
next_resp = prompt2response(next_prompt)
print(next_resp.content)

# print chat history
list(map(print,next_resp.chat_log()))

License

This package is licensed under the MIT license. See the LICENSE file for more details.

Features

• TODO

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.