In [ ]:

Welcome to My Version of Brandon’s Pandas Tutorial

https://github.com/brandon-rhodes/pycon-pandas-tutorial

The first instance of this tutorial was delivered at PyCon 2015 in Montréal, but I hope that many other people will be able to benefit from it over the next few years — both on occasions on which I myself get to deliver it, and also when other instructors are able to do so.

If you want to follow along with the tutorial at home, here is the YouTube recording of the 3-hour tutorial at PyCon itself:

https://www.youtube.com/watch?v=5JnMutdy6Fw

To make it useful to as many people as possible, I hereby release it under the MIT license (see the accompanying LICENSE.txt file) and I have tried to make sure that this repository contains all of the scripts needed to download and set up the data set that we used.

Quick Start

If you have both conda and git on your system (otherwise, read the next section for more detailed instructions):

Once you have unpacked the ZIP file, download the following four IMDB data files and place them in the tutorial’s build directory:

  • ftp://ftp.fu-berlin.de/pub/misc/movies/database/actors.list.gz
  • ftp://ftp.fu-berlin.de/pub/misc/movies/database/actresses.list.gz
  • ftp://ftp.fu-berlin.de/pub/misc/movies/database/genres.list.gz
  • ftp://ftp.fu-berlin.de/pub/misc/movies/database/release-dates.list.gz

Notes

Prolog:

There are 4 parts to this workshop:

  1. Intro to pandas and numpy
  2. Exploring the dataset
  3. Feature engineering
  4. Builgin a model

Intro

  • Why are we learining pandas and numpy? (Learning ML-algorithms is like CS-algorithms. Most of the work is coding!)

In [ ]:


In [ ]: