Skip to content

ChatResearch is an indispensable tool that could be a lifesaver for graduate students.

License

Notifications You must be signed in to change notification settings

cauliyang/ChatResearch

Repository files navigation

ChatResearch

pypi python version license precommit black Ruff

ChatResearch is a tool that uses OpenAI's GPT-3 to accelerate your research. It provides several features such as generating summary papers, fetching and summarizing papers from arxiv and bioarxiv, generating responses for review comments, and more.

❓ Why ChatResearch

There have been countless projects and research initiatives in the field of ChatGPT, but none have fulfilled my particular needs. Consequently, I have made the decision to develop my own project. I am committed to continuously refining and improving this project. Your support and ratings are greatly appreciated, so please do not hesitate to leave a star. I also welcome any suggestions, proposals, and pull requests.

🚀 News

  • Multi thread and Asynchronous support has been added.
  • GIFs have been added for documentation.
  • Config can now be used to set key.
  • Output for Latex and PDF is now supported.

🔖 TODO

  • Multi thread and Asynchronous support
  • Add GIF for document
  • Use config to set key
  • Support output for Latex and PDF
  • Fine-tune prompt
  • Generate Image
  • Add RSS support for multiple journal
  • Revise help message
  • add classic paper as examples

🌟 Features

  • Chat Config: Generate chatre.toml in current working directory or set environment variable OPENAI_API_KEY.
  • Chat Paper: Fetch or summary paper from local or arxiv with specified query, research fields, and language.
  • Chat Biorxiv: Fetch and summary paper from bioarxiv with specified category, filter keys, and language.
  • Chat Reviewer: Generate summary paper with specified research fields and language.
  • Chat Response: Generate response for review comment with specified language.
  • Markdown, PDF, and Latex

⚙️ Installation

pip install chat-research

:face_with_monocle: Usage

Chat Config

chatre config

It will generate chatre.toml in current working directory. Otherwise, setting environment variable OPENAI_API_KEY is another way to config API KEY. Also, chatre config set OPENAI_API_KEY=sk-key will create configuration file in current file directory and set your key as sk-key.

[OpenAI]
OPENAI_API_KEYS = [ "sk-key1", "sk-key2",]

[Gitee]
api = "your_gitee_api"
owner = "your_gitee_name"
repo = "your_repo_name"
path = "files_name_in_your_repo"

Chat Paper

Chat Biorxiv

Chat Response

Chat Reviewer

> chatre reviewer -h
usage: chatre reviewer [-h] --paper-path  [--file-format] [--review-format] [--research-fields] [--language]

optional arguments:
  -h, --help          show this help message and exit
  --paper-path        path of papers
  --file-format       output file format (default: txt)
  --review-format     review format
  --research-fields   the research fields of paper (default: computer science, artificial intelligence and reinforcement learning)
  --language          output language, en or zh (default: en)

👊 Develop Toolkit

License

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

Contributors

  • Yangyang Li

Alt

👏 Acknowledgement

About

ChatResearch is an indispensable tool that could be a lifesaver for graduate students.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages