Simple example of autonomous research ran in parallel for an eventual cluster compute solution from my Aetherius Ai Assistant project. Mainly meant for others to use in their projects.
Websearch Implemented!
Change Log
Version 0.12
-Research GPT now decides for itself if it should conduct a web search or not. Also serves as a showcase on how to implement simple tools.
-Redid the way if conditions were done to avoid multi-threading issues.
-Added parameter for limiting answers per task.
Side Project of the Aetherius Ai Assistant
Aetherius GitHub: https://github.com/libraryofcelsus/Aetherius_AI_Assistant
-
Install Git: https://git-scm.com/ (Git can be skipped by downloading the repo as a zip file under the green code button)
-
Install Python 3.10.6, Make sure you add it to PATH: https://www.python.org/downloads/release/python-3106/
-
Open the program "Git Bash".
-
Run git clone: git clone https://github.com/libraryofcelsus/Research_GPT.git
-
Open CMD as Admin
-
Navigate to Project folder: cd PATH_TO_RESEARCH_GPT_INSTALL
-
Create a virtual environment: python -m venv venv
-
Activate the environment: .\venv\scripts\activate (This must be done before running Research GPT each time, using an IDE like PyCharm can let you skip this.)
-
Install OpenAi's packages with pip install openai
-
Copy your OpenAI api key to key_openai.txt (https://openai.com/blog/openai-api)
-
Copy you Bing api key to key_bing.txt (https://www.microsoft.com/en-us/bing/apis/bing-web-search-api)
-
Run Research GPT with **python Research_GPT.py