All notebook material: https://github.com/INRIA/scikit-learn-mooc/

A few different ways are available:

• Launch an online notebook environment using
• Browse website generated with Jupyter Book

The tutorials will require the following packages:

• python>=3.6
• jupyter
• scikit-learn
• pandas
• matplotlib
• seaborn
• plotly
• jupytext (required only for contributors)

We provide both requirements.txt and environment.yml to install packages.

You can install the packages using pip:

$ pip install -r requirements.txt

Alternatively, you can create an scikit-learn-tutorial conda environment by executing:

$ conda env create -f environment.yml

then activate the environment with:

$ conda activate scikit-learn-tutorial

You can also update your current environment, instead of creating a new environment, using:

$ conda env update --prefix ./env --file environment.yml  --prune

The source files, which should be modified, are in the python_scripts directory. The notebooks are generated from these files.

This repository uses Jupytext to display Python files as notebooks. Saving as Python files facilitates version control.

When jupytext is properly connected to jupyter, the python files can be opened in jupyter and are directly displayed as notebooks

With jupyter notebook

Once jupytext is installed, run the following command:

jupyter serverextension enable jupytext

With jupyter lab

To make it work with "jupyter lab" (instead of "jupyter notebook"), you have to install nodejs (conda install nodejs works if you use conda). Then in jupyter lab you have to right click "Open with -> notebook" to open the python scripts with the notebook interface.

To update all the notebooks:

$ make

If you want to generate a single notebook, you can do something like this:

$ make notebooks/02_numerical_pipeline_scaling.ipynb

• OVH Binder

• GESIS Binder

• GKE Binder