project address

https://github.com/binary-husky/chatgpt_academic

configuration method

download item

Enter the Github link and download the project

install dependencies

• pip download

cd to enter the project directory, use the command python -m pip install -r requirements.txt
(ps: downloading with pip may cause dependency problems due to the old version, just re-download/update the corresponding package according to the relevant error report)

• Conda download (need to create a new environment, which is more troublesome)

conda create -n gptac 'gradio>=3.23' requests
conda activate gptac
python3 -m pip install mdtex2html

openAI API acquisition and proxy configuration

openAI API acquisition

1. Enter openai official website https://platform.openai.com/account/api-keys
2. Click +Create New secret key to get the keys and copy them (note that this key cannot be viewed again, if you forget it, you can only regenerate it)
3. Enter the project config.pyfile and modify API_KEY to your own KEY

API_KEY = "刚才复制的KEY"

proxy configuration

The ladder used before only configured the browser side, here is the configuration of the local client.

1. Download clash and configure the proxy according to the software instructions
2. Enter config.pythe file and modify:
   1. USE_PROXY = True
   2. proxies = { "http": "http://localhost:7890", "https": "http://localhost:7890", }
      

run project

After entering the project path python main.py, it will automatically open the link
PS: Maybe there will be some errors at this time. After copying, go to Baidu to check the solution. Generally, the version of the package does not match, just download the corresponding one.