joelgarciajr84 / Bayesnote

A frictionless integrated platform for notebook

Bayesnote

Bayesnote is a notebook orchestration platform for Jupyter notebook:

• Orchestrate notebooks on different clusters within 5 minutes
• One-button access to remote Jupyter notebook
• Run Jupyter notebooks with dependency and retry
• View output notebook with one-button

Why:

• No learning curve. There is no abstraction to learn, no docker/Kubernetes knowledge is required.

• No operation cost. There is no extra database to install, no ports to open. It can install itself into clusters without hassles.

Screenshots

Installation

1. Download for your operation system https://github.com/Bayesnote/Bayesnote/releases
2. Open terminal && Assign permissions chmod 777 bayesnote
3. Go the download folder && Run in the terminal ./bayesnote
4. MacOS user only: Go to "Settings"-> "Security & Privacy"-> "Allow" -> "Open"

The web browser would open at localhost:8080.

Tutorial

Hosts:

0. Make sure Jupyter notebook is already installed on the host
1. Add the host you want to run your notebook by SSH credentials
2. Open remote notebook with one-click and the browser will open Jupyter notebook on a random port.
3. Start writing code print("Hello bayesnote")

Flows:

1. Create a new flow. Add depdency in next.
2. Click Start.
3. View the notebook by click show. The browser would open a new tab to show output notebook.

How to view output:

• Click Show -> Click Notebook

Ariticles:

Orchestrate Jupyter Notebooks in 5 minutes

Note: This project is adding new features fast.

Note: I am working on this project full-time. Let me know what do you think. Questions, comments, feature requests, and pull requests welcome. welcome.