A Jupyter Notebook-based workshop on data visualization and machine learning basics created and led by Sami Friedrich and Alex Nevue. This workshop was part of a peer-instructed Python course in 2018 at Oregon Health and Science University organized by Dr's Stephen David and Brad Buran. Teacher, student, and answer notebook versions are derived from the base notebook file.
This workshop demonstrates some of the neat ways to explore and display data using matplotlib
, seaborn
and bokeh
. Then it introduces scikit-learn
, and the power of classifiers to make predictions.
The dataset used throughout this workshop is from a study published in 2016 by Olcowicz et al. titled Birds have primate-like numbers of neurons in the forebrain. It is predominantly composed of brain cell counts from multiple brain areas of different bird species.
- Understand the unique benefits of
matplotlib
,seaborn
, andbokeh
data visualization packages - Enrich data storytelling using
seaborn
to encode additional variables and improve readability in plots - Create engaging, interactive & responsive plots using
bokeh
- Implement the 4-step method to instantiate, fit, predict, and evaluate models using
scikit-learn
- Generate predictive classifiers using a k-nearest neighbors approach
- Introduction
- Loading the data
- Data maniupulation
- Data visualization using
matplotlib
- Data visualization using
seaborn
- Data visualization with
bokeh
- Machine Learning Basics with
scikit-learn
- Wrap-Up
Sami Friedrich
Alex Nevue
Brad Buran provided valuable feedback that greatly improved this workshop.