These are the slides, demo files, and datasets used for the "Machine Learning for Humans" talk given at the Los Alamos National Laboratory. See the "Resources" silde appended near the end of the presentation for links.
-
ml_for_humans.pdf: the slides -
demos/demo1.py: linear regression, printing coefficients -
demos/demo2.py: linear regression, cross validation -
demos/demo3.py: multi-model comparison -
demos/util.py: contains small utility function to read CSV files -
demos/cold_rocks.csv: dataset for restaurant prediction demo (Weka) -
demos/hw_data.csv: partial and obfuscated Kraken data -
demos/mpg_data.csv: classic car mileage dataset
Please note that all demos require the installation of scikit-learn, and are intended to be ran with Python 3.
Find me on Twitter, LinkedIn, GitHub, and at my blog.
This work, including source code, is licensed a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. The cold_rocks.csv data was created entirely by myself and may be considered public domain. The mpg_data.csv file was derived from the UCI Machine Learning Repository and is subject to any restrictions placed on it by UCI.
The hw_data.csv is provided gratis but may only be used as provided and cannot be redistributed, cited, or modified regardless of attribution or intent. You may use it for running demo3.py.