MindMaze
Algorithmically generated maze animations

A randomized maze generator and animator. Supports run-time selection from generation algorithms including: Kruskal's, sidewinder, binary tree, and depth first search. Solutions can be found using depth first search, breadth first search, and A* search with multiple heuristics.

Originally designed to compare efficiency of algorithms for generating and solving mazes. Animation capability was added later to provide a visualization.

Coded in Python with libraries including numpy and matplotlib.


Source Code

Echoes
A music based deep learning application

A WIP project used as a documentation and exploration of my learning process for ML and deep learning.

Involves organization and manipulation of data in the form of audio files using numpy and pandas. Visualizations of the data were created with librosa, matplotlib, and seaborn.

Multiple models are trained on a data set of Bach compositions using SciKit learn and Pytorch.


Source Code

UMN Projects
Computer science curriculum projects

  • An automated drone program which utilizes computer vision, image processing, design templates, and 3D motion to search an area for a target and send help to the location (C++)
  • A concurrent producer-consumer banking scenario utilizing multithreading, semaphores, sockets and signals in order to achieve high performance (C)
  • A tautology checker and Lisp interpreter written with functional programming (OCamL)
  • A Battleship game (Java) and a player v. computer version of snake (Python)

Source Code