A Gentle Introduction to Machine Learning with Jupyter Notebooks

This is a collection of short jupyter notebooks intended to give the absolute beginner a taste of machine learning using Python and the Scikit Learn library.

Installation Requirements

You must have the following installed:

• Python 3
• Pipenv
• Jupyter Notebook (e.g. via Anaconda)
• Git

You can verify that these were correctly installed by running the following and comparing the output. What matters most is that the python version starts with a 3, the rest can vary slightly and that is O.K.

> python --version
Python 3.6.3
> pipenv --version
pipenv, version 8.3.2
> jupyter notebook --version
5.0.0
> git --version
git version 2.15.1 (Apple Git-101) 

If you do not get a version number, you should troubleshoot your installation. If the python version number is 2, you need to install python 3 (you can install both versions concurrently).

Setup

Change directories to the directory you wish to be in and clone the repository via git using the following command:

git clone https://github.com/rfblue2/gentle-intro-to-ml.git
cd gentle-intro-to-ml

We will use pipenv to set up our environment.

pipenv --three
pipenv install

Running the notebook

We start the jupyter notebook within our environment by running the following command:

pipenv run jupyter notebook

This should redirect you to your browser but if not, you can type localhost:8888 or whatever number the output tells you to go to, and it will show you a nice interface with all the files in it including the .ipyb files, which you can double click to open.

Further Reading

If you want to experiment more, I highly recommend getting a copy of Hands-On Machine Learning with Scikit-Learn and Tensorflow by Aurelien Geron.