🤖🧪 AutoResearcher

⚡ Automating scientific workflows with AI ⚡

What is AutoResearcher?

AutoResearcher is an open-source Python package that leverages AI models and external knowledge sources to automate scientific workflows. Designed to help researchers and scientists accelerate their research process and increase efficiency, AutoResearcher is a powerful tool for the modern scientific community.

Please note that the project is currently in its early prototype stage and under active development. Its present functionality is limited to conducting literature reviews, but the ultimate goal is to create a tool capable of driving scientific discovery on autopilot.

If this vision excites you, we invite you to contribute to the project. Start by joining our Discord server and discussing your ideas with our community.

Documentation

Documentation for the package is available here.

Installation

Install the package using pip:

pip install autoresearcher

Setting Environment Variables

Before using the package, you need to set the following environment variables:

• OPENAI_API_KEY: Your OpenAI API key for accessing the GPT-based AI models.
• EMAIL: An email address of your choice (used to identify your API requests for getting citations).

You can set the environment variables in your operating system or in your Python script using the os module:

import os

os.environ["OPENAI_API_KEY"] = "<your_openai_api_key>"
os.environ["EMAIL"] = "<your_email>"

Replace <your_openai_api_key> and <your_email> with your actual API key and email address.

Usage

1. Import the literature_review function from the package:

from autoresearcher import literature_review

2. Set your research question as a string:

research_question = "What is the best way to train a neural network?"

3. Create a literature_review instance with your research question and execute it:

researcher = literature_review(research_question)

You can also pass an output file name as a .txt file:

researcher = literature_review(research_question, output_file="my_literature_review.txt")

This will generate a literature review based on the research question.

Also, you can run it in one command:

python run_autorsearcher.py --research_question "<your_research_question>" --output_file "<your_output_file>"

4. To use GPT-4o instead of the default GPT-4o-mini, set the use_gpt4 parameter to True:

researcher = literature_review(research_question, use_gpt4=True)

Note that using GPT-4o may result in higher costs and potentially longer processing times.

Contributing

We welcome contributions! Feel free to submit issues or create pull requests. Together, let's revolutionize science! 🚀

License

This project is licensed under the MIT License. See the LICENSE file for details.

Made with ☕ by @eimenhamedat