Trevor Brunnenmeyer, PhD
I use the skills developed earning a PhD in Physics to solve Data Science problems.
View My LinkedIn Profile
Portfolio of Select Data Science and Machine Learning Projects
Hiscores Analysis for Runescape’s Fresh Start Worlds - Part 2
An analysis of Runescape Fresh Start Worlds players in the main game after the conclusion of Fresh Start Worlds. The data for this project was scraped from the Runescape Hiscores using Beautiful Soup and then analyzed in Python using pandas, matplotlib, and Seaborn. Finally, I did a k-means clustering analysis using sklearn and visualized the cluster centers using Seaborn.
Hiscores Analysis for Runescape’s Fresh Start Worlds
An analysis of the top players of the Runescape Fresh Start Worlds’ competitive period. The data for this project was scraped from the Runescape Hiscores using Beautiful Soup and then analyzed in Python using pandas, matplotlib, and Seaborn.
PokéGAN: A Neural Network for Making Fake Pokémon Sprites
Used Keras to build a Generative Adversarial Network (GAN) to generate new Pokémon sprites. In the project I experimented with different depths of CNNs and various ways to train the GAN.
Generating Handwritten Digits
Used the MNIST handwritten digits to train a GAN to generate fake handrwitten digits. In this project I learned to use Keras to build Convolution Neural Networks while using a well curated data-set.
Set Tutorial and Lab Teaching Assignments
Each quarter we would have an hour plus meeting to assign TAs to tutorial and lab sections. This process was slow and inequitable, as the TAs who spoke loudest and quickest often got the better assignments. I took the initiative to collect ranked preferences from each TA and then wrote code in Python to optimize the assignments based on the groups preferences. This reduced the required meeting time by 50% and improved the equity in assignments.
Page template forked from evanca