Skip to main content

The Bridge from Thoughts to Well-Crafted Jupyter Notebook

Project description

Unit Tests PyPI PyPI - Python Version GitHub

🚀 Notebook Copilot: Turn Your Thoughts Into a Polished Notebook at Record-Speed with AI.

Welcome to Notebook Copilot, your next-generation tool for Jupyter Notebooks. Inspired by GitHub Copilot, Notebook Copilot is designed to help data scientists and engineers in developing professional, high-quality notebooks. It's like having your personal AI-powered assistant that helps you navigate through the Jupyter universe, seamlessly generating code and markdown cells based on your inputs.

Imagine not having to start with a blank notebook every time. Sounds dreamy, right?

v1 Demo

https://github.com/talperetz/notebook-copilot/assets/11588598/02e2b030-adb3-42f2-bd6d-4a8451e3831f

Features

  • 🚀 GPT Based Generation: Notebook Copilot employs advanced GPT instances for precise and efficient code generation.
  • 💻 Integrated with Any Notebook Environments: Seamless access within Jupyter Notebook and other popular platforms, boosting your productivity.
  • 🧩 Automatic Context Retrieval: Understands the full context of your notebook, ensuring consistent and relevant code generation.
  • 🔑 Bring Your Own Key: Flexibility to use your own OpenAI key for personalized code generation and optimal results.
  • 🆓 Free and Open Source: Everyone can benefit from Notebook Copilot. It's our contribution to the coding community, aiming to make coding accessible, efficient, and fun.

Quickstart

!pip install notebook_copilot
%load_ext notebook_copilot
%copilot

Walk-through

  1. Get an OpenAI API Key
  2. Install Notebook Copilot directly from PyPI:
pip install notebook_copilot
  1. Load the Notebook Copilot extension in your Jupyter notebook:
%load_ext notebook_copilot

# Optional: If you don't have OPENAI_API_KEY set in your environment, you can set it here
from getpass import getpass
import os
os.environ["OPENAI_API_KEY"] = getpass("Enter your OpenAI Key: ")
  1. Start using Notebook Copilot Magic Functions in your notebook ↓

✨ Magic Functions

🪄 Enter Assistant Mode and let Copilot continuously generate professional code and markdown cells for you.

%copilot

✍️ Leverage AI to create the next cell from your comments. It's like having a conversation with your notebook.

%%code 
# Plot the confusion matrix for each model
# Plot the precision-recall curve for Catboost

📘 Automatically generate markdown cells to explain the code in the current cell. Your code is now not only functional but also well-documented.

%%explain
# some code to explain…

⚡ Improve the time complexity of the code in no time. Copilot will generate an alternative code cell that is optimized for speed.

%%optimize
# a code cell…

🎨 1 plot >= 1000 data rows. Visualize your data with a single line of code.

%%visualize
# a code cell…

Roadmap

  • Copilot Magic Function: Continues the notebook for you, generating professional code and markdown cells, making blank notebooks a thing of the past.
  • Generate Magic Function: Turn Your Comments into Code
  • Explain Magic Function: Generate Markdown Cells that Explain Your Code
  • Optimize Magic Function: Generate an Aleternative Code Cell that is Optimized For Speed / Simplicity
  • Visualize Magic Function: Generate a Cell that Visualize Your Data
  • Speed improvenents
    • Support parallel cell generation
  • Update underlying strategy and prompts
  • Support more llm providers
    • Starcoder
    • Anthropic

Contributing

We appreciate all contributions. If you're planning to contribute back bug-fixes, please do so without any further discussion. If you plan to contribute new features, utility functions, or extensions to the core, please first open an issue and discuss the feature with us.

License

Notebook Copilot is MIT licensed, as found in the LICENSE file.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

notebook_copilot-0.2.0.tar.gz (14.2 kB view details)

Uploaded Source

Built Distribution

notebook_copilot-0.2.0-py3-none-any.whl (15.2 kB view details)

Uploaded Python 3

File details

Details for the file notebook_copilot-0.2.0.tar.gz.

File metadata

  • Download URL: notebook_copilot-0.2.0.tar.gz
  • Upload date:
  • Size: 14.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.13

File hashes

Hashes for notebook_copilot-0.2.0.tar.gz
Algorithm Hash digest
SHA256 110cc3584a629f86f639e26cff39ebc19508110d846b85fbdea99c9a396033e2
MD5 0afa9d75e9ca93c97e3478da8e71420b
BLAKE2b-256 7c898efcd02cd00a578a7e76094d007eb729b16419d6bd2a8c97f8feb47c87e2

See more details on using hashes here.

File details

Details for the file notebook_copilot-0.2.0-py3-none-any.whl.

File metadata

File hashes

Hashes for notebook_copilot-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 85f53550af5d05d5c04f817d0db41e059a442e663bd3035559c5f4e238f8edcf
MD5 788b4863968a1f04c735cea38de9c652
BLAKE2b-256 1296ffcdeab9ddad34752566a88b1e201c147570c64b5c3044c540dc2ad2c723

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page