A learning by doing workshop for learning the basics of the Pandas library.
The main prerequisite for this workshp is Jupyter with NumPy and Pandas installed (or some other way of using a Jupyter notebook). There are many services that provide notebooks for free. If you do not want to install or run on your local computer, you can follow the entire workshop using one of the following free Jupyter services:
There are multiple ways of installing Jupyter and for this particular workshop I recommend one of two ways: by using a docker container or by using the Anaconda distribution.
NOTE: If you are using one of the notebook solutions in the cloud, you have to upload the notebook, img folder and data folder to the cloud provider (or check it out from git on the cloud providers notebook) This workshop is designed around using docker and a local computer mainly.
Install docker on your local computer, if this is not already installed. You can download docker from here: Download Docker
After installation, pull the latest Jupyter image and start it on your computer. If you use the command line then you can do this by typing the following command:
Windows:
docker run --rm -p 8888:8888 -e JUPYTER_ENABLE_LAB=yes -v "%cd%":/home/jovyan/work jupyter/datascience-notebook
OSX / Linux:
docker run --rm -p 8888:8888 -e JUPYTER_ENABLE_LAB=yes -v "$PWD":/home/jovyan/work jupyter/datascience-notebook
A container using the latest jupyter/datascience-notebook will then be started and can be stopped by pressing Ctrl-C in the console (or closing the console). In the console a link is printed that you can use to access the notebook. This link looks like this: http://127.0.0.1:8888/?token=6519a549112e9191d6485a5360980635442c8ec9e03f4924 where the token is different.
You now have a running Jupyter Notebook instance which runs as a Docker container and you can work on notebooks using your favorite browser. The notebooks mounts up your current working directory as the folder work in Jupyter.
Anaconda is a Python distribution that contains a lot of packages that often is not present in PiP. You can download an installer and install it locally by downloading the installer here.
Much of the content in this workshop is based upon a book that is available for free: The Python Data Science Handbook written by Jake VanderPlas. This is one of the best books for learning Jupyter, NumPy and Pandas. The book is free and available for reading at the previous link. There is also a print copy that you can buy from Amazon here: Python Data Science Handbook
The book will not be used directly in the workshop. It is a highly recommendend read if you like to learn by yourself outside of the workshop.
