# Project¶

Our course project will be a gentle introduction to a topic not directly covered in class. You will introduce the topic to your fellow classmates. The topic can be a subtopic of a topic we covered in class or something not covered in class at all.

Since there are roughly 25 students in this class, and every student will give a 10 minute presentation, our projects will be due by the end of our scheduled final exam minus 250 minutes of class time.

## Deliverables¶

You will

create a Jupyter notebook,

deliver a 10 minute presentation using 1., and

provide a compliment of every other students project

The Jupyter notebook is intended to help your classmates follow along your introduction to your chosen project idea. Your introduction should provide context to the (sub)topic and provide at least two reasonable examples.

Your notebook and compliments will submit via GitHub in separate Jupyter notebooks. I’d like to automate (as much as possible) the distribution of everyone’s compliments, so there will be strict, and as of yet undetermined, structure to follow for the compliments.

## Project Ideas¶

Below is a list of project ideas and/or resources, including some ideas that have already been taken (those with a 🚫). The not taken project ideas are up for grabs. You may also use the list to help you think of other project ideas not listed here.

If I have/find some good resources, I’ll list them underneath each project idea.

- 🚫 Python packages (computery)
Original tools setuptools

Or newer tools Alex Kyllo’s post Easy Python Package Publishing with Poetry

Python package poetry

Don’t forget documentation; use Sphinx

Don’t forget continuous integration.

- yada yada yada plots (statsy, computery)
Tips for contour plots in Python’s Matplotlib from Jake Vanderplas.

🚫 Louis Tiao has a nice post about visualizing and animating optimization algorithms with matplotlib.

Making a plot to highlight some of points made in the following links would be great! Sebastian Ruder has a fairly popular blog post which provides an overview of some of the more popular gradient descent optimization algorithms, An overview of gradient descent optimization algorithms. John Chen provided a follow up post An updated overview of recent gradient descent algorithms, which includes some more mordern methods.

- git (computery)
Read either git book or Getting Git Right, but stick to just one at first

🚫 Workflows, stick to either Centralized or Feature Branch.

🚫 Oh Shit, Git!?! “Git is hard: screwing up is easy, and figuring out how to fix your mistakes is fucking impossible.”

🚫 Does NBA draft pick order correlate with players’ career success?

- Gradient Descent methods (mathy)
Sebastian Ruder has a fairly popular blog post which provides an overview of some of the more popular gradient descent optimization algorithms, An overview of gradient descent optimization algorithms.

more resources to come…

- Linear Algebra on a computer (mathy, computery, advanced)

- Deep Learning (mathy, computery, advanced)

- Association Rules (statsy)
A Probabilistic Comparison of Commonly Used Interest Measures for Association Rules; pay attention to at least Support, Confidence, and Lift.

🚫 Dig deeper on the AITA subredit data.