pycharm 导入Auto-GPT（墙内）

引言

AutoGPT 是让 GPT-4 自动化的驱动，通过设定5个目标，自动搜索数据并喂给GPT，同时分析返回结果。

准备工作

在接入 AutoGPT 之前，你需要完成以下几个步骤：

1. 安装 PyCharm
2. 科学上网：clash

pip install openai-autogpt

安装Auto-GPT

• 打开终端执行

git clone https://github.com/Torantulino/Auto-GPT

• 打开pycharm 导入项目

• 将.env.template 重命名为 .env

• 修改.env 的配置文件

OPENAI_API_KEY=  #获取地址 https://platform.openai.com/
PINECONE_API_KEY= #获取地址 https://www.pinecone.io
GOOGLE_API_KEY= #获取地址 请参考 https://zhuanlan.zhihu.com/p/621746507

关键 设置代理

• • 打开 tiktoken 下面的load.py ，然后修改 read_file 方法，增加代理

def read_file(blobpath: str) -> bytes:
    if not blobpath.startswith("http://") and not blobpath.startswith("https://"):
        try:
            import blobfile
        except ImportError:
            raise ImportError(
                "blobfile is not installed. Please install it by running `pip install blobfile`."
            )
        with blobfile.BlobFile(blobpath, "rb") as f:
            return f.read()
    # avoiding blobfile for public files helps avoid auth issues, like MFA prompts
    proxies = {
        'http': 'http://127.0.0.1:7890',
        'https': 'http://127.0.0.1:7890'
    }
    return requests.get(blobpath, proxies=proxies).content

• • 打开 openai 下面的 api_requestor.py,修改 request_raw 方法为

def request_raw(
      self,
      method,
      url,
      *,
      params=None,
      supplied_headers: Optional[Dict[str, str]] = None,
      files=None,
      stream: bool = False,
      request_id: Optional[str] = None,
      request_timeout: Optional[Union[float, Tuple[float, float]]] = None,
  ) -> requests.Response:
      abs_url, headers, data = self._prepare_request_raw(
          url, supplied_headers, method, params, files, request_id
      )

      if not hasattr(_thread_context, "session"):
          _thread_context.session = _make_session()

      try:
          proxies = {
              'http': 'http://127.0.0.1:7890',
              'https': 'http://127.0.0.1:7890'
          }

          result = _thread_context.session.request(
              method,
              abs_url,
              headers=headers,
              data=data,
              files=files,
              stream=stream,
              proxies=proxies,
              timeout=request_timeout if request_timeout else TIMEOUT_SECS,
          )
      except requests.exceptions.Timeout as e:
          raise error.Timeout("Request timed out: {}".format(e)) from e
      except requests.exceptions.RequestException as e:
          raise error.APIConnectionError(
              "Error communicating with OpenAI: {}".format(e)
          ) from e
      util.log_debug(
          "OpenAI API response",
          path=abs_url,
          response_code=result.status_code,
          processing_ms=result.headers.get("OpenAI-Processing-Ms"),
          request_id=result.headers.get("X-Request-Id"),
      )
      # Don't read the whole stream for debug logging unless necessary.
      if openai.log == "debug":
          util.log_debug(
              "API response body", body=result.content, headers=result.headers
          )
      return result

最后

展示一下成果

PS：之后在遇到问题，请直接问 gpt-4

题外话

最近ai 特别火，拿Stable-Diffusion 创作一副，其他的太过XX，就不放了