/pandas_for_everyone

Repository to accompany "Pandas for Everyone"

Primary LanguageJupyter NotebookMIT LicenseMIT

Pandas for Everyone

Repository to accompany "Pandas for Everyone".

If you have gone through the book, an Amazon review would be much appreciated! My mom would too :)

Setup

The easiest way to get everything you need to the tutorial is to install anaconda

You can download and install it here: https://www.continuum.io/downloads

To download just the data, see the Data section below. Otherwise you can choose to clone this repository, or click the "Clone or Download" link above and clicking Download Zip

Install seaborn for plotting

conda install seaborn

Install all the packages used in the book

There is an error in the preface of the book for installing packages. I am leaving this section here in the README to have an updated list of packages and installation instructions

(Optional) Create a Virtual Environment

You can choose to create a virtual envirionment for the packages used in the book, so it doesn't clash with other packages you plan to use later on.

# create a virtual environment named "book" using python 3.6
conda create -n book python=3.6

# activate the environment
# so all installed packages will go in there and not mess up your base python environment
source activate book

Install the packages

Whether you decited to create a virtual environment or not, you can install the packages with the below commands. If you did use virtual environments, remember to source activate book before you follow along with the book so the packages you installed can be loaded.

conda install pandas xlwt openpyxl seaborn numpy ipython jupyter statsmodels scikit-learn regex wget odo numba
conda install -c conda-forge pweave # you don't really need this package, it was used to build and create the book
conda install -c conda-forge feather-format
pip install lifelines pandas-datareader

Teaching Slides

For those instructors who are using the teaching slide deck version of the book. Each chapter is split into it's own slide deck. There are multiple versions for each chapter.

  1. Jupyter notebook (ipynb)
  2. PDF
  3. HTML

The slides are created using Damian Avila's RISE Jupyter/IPython Slideshow Extension. Thus, you can choose to install the RISE extension and live render and display the Jupyter notebooks (ipynb). Since each chapter is a Jupyter notebook at heart, the conversions to PDF and HTML are performed using

jupyter nbconvert --to slides your_talk.ipynb --post serve

More about useage ange converting to the PDF can be found on the RISE documentation page on useage.

No Powerpoint (.ppt/.odp)

RISE's back end uses reveal.js. Unfortunately there is no way to go from a reveal.js presentation to powerpoint. Having said that, if there's a way we can jerry-rig something together using the the given capabilties of RISE and reveal.js please let me know.

Data

You can choose to just download the datasets by using Minhas Kamal's DownGit by clicking the link here

Ongoing list of data references:

  1. Gapminder: https://github.com/jennybc/gapminder/
  2. Survey: Comes from the Software-Carpentry SQL lesson
  3. Ebola: www.github.com/cmrivers/ebola

Links to teaching sessions

I've taught out of the book while I was writing it. Here you can find the various tutorials and workshops I've taught (pre and post when the book was officially published). You can also checkout my talks page for other things not completely on Pandas.

Tables URL Video
Online Live Training https://github.com/chendaniely/2017-12-04-pandas_live, https://github.com/chendaniely/2018-05-pandas_live, https://github.com/chendaniely/2018-06-pandas_live
Whirlwind tour of Python https://github.com/chendaniely/2017-10-26-python_crash_course
SciPy 2017 Pandas Tutorial https://github.com/chendaniely/scipy-2017-tutorial-pandas https://www.youtube.com/watch?v=oGzU688xCUs
PyData Carolinas 2016 Tutorial https://github.com/chendaniely/2016-pydata-carolinas-pandas https://www.youtube.com/watch?v=dye7rDktJ2E

Other random goodies