Jupyter Notebook is an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and explanations. In fact, it’s an interactive IDE, allowing you to share your work with anyone, collaborate and review your Python code.
It’s a web-based tool, though you can run it on your local machine (just as I do). You could check how it works by opening and running through any so called notebook (a page file with .ipynb extension, at the Jupyter Project site https://try.jupyter.org/
Running your local Jupyter instance is pretty simple. After installation, just open command line in directory were you’d like to store/open your .ipynb files and hit command:
That’s it! A browser page «http://localhost:8888/tree» will shortly pop up displaying your own Jupyter local instance and you’re ready to rock.
Jupyter is widely used in Machine Learning and Data Science areas as it has not only great UI and store your computations as if you’d have in paper notebook, but also data cleaning and transformation, numerical simulation, statistical modeling, etc.
I use it to run scikit-learn, pandas, numpy and matplotlib libraries, all great for the Data Mining/Analysis and Machine Learning tasks.
Here is also «A gallery of interesting Jupyter Notebooks«, worth to visit and check it out to see loads of the examples where Jupyter Notebook can be useful.