Will
initial proposal
----------------------------------------------
--- Tutorial proposal form ---
----------------------------------------------
Name : William Linkmeyer
Tutorial title : Machine Learning in Python
Desired credits : 1 to 3 — subject to your discretion
Tutorial description: (appropriate for the registrar's permanent record)
Write an algorithm in Python 3 that takes a non-arbitrary set of data
and outputs a relevant and coherent prediction on said data set.
(ex: giving a program designed to estimate the value of a
house a set of data on a house, one should expect that program to
output an estimated value of that house)
What exactly do you want to study?
Be as explicit as you can, including a schedule if possible.
The fundamentals of Python (X.x) and scikit-learn.
Conceptual machine learning techniques and their use in scikit-learn.
How does this relate to your plan and/or other course work?
I plan for my PLAN to be on Machine Learning and Artificial
Intelligence and their use with the English Language.
A tentative example would be a program capable of, given all of
Shakespeare’s plays as input, outputting a Shakespearean-style play,
and continuously learning from future input in an unsupervised manner.
Considering my background and my PLAN, I believe Machine Learning in a
scripting language — Python — using an API (in lieu of writing my own from
scratch) — scikit-learn — is a good starting point for me. This would teach
me the fundamentals of Python in a depth greater than an introductory course
as well as conceptual machine learning in a depth far less than the theory
and practice of machine learning algorithms would demand of me
What resources have you identified?
(e.g. books, articles, websites, experience, ...)
Book:
Machine Learning in Python (V. 1), Michael Bowles, 2015.
Websites, Documentation, and Videos:
scikit-learn API: (http://scikit-learn.org/stable/)
scikit-learn documentation: (http://scikit-learn.org/stable/documentation.html)
“Machine Learning with Text in scikit-learn”, Kevin Markham, From PyCon 2016:(https://www.youtube.com/watch?v=ZiKMIuYidY0)
“Machine Learning with Scikit Learn”, Jake VanderPlas, From PyData 2015: (https://www.youtube.com/watch?v=HC0J_SPm9co)
Experience:
Four years of unstructured and independent programming mostly using C & OpenGL.
What will be the gradeable products, and on what schedule?
(e.g. projects, programs, papers, tests, ...)
At least two programs—with the possibility of one being
a distinctly newer and more functional version of a previous
program—will be written for you to grade.
Programs should be available at the latest by the middle of the semester and by the end of the semester.
The number of programs should be subject to their complexity.
Frequent and open communication—at least twice a week—about my progress will be provided by myself.
