Posts by Collection

portfolio

PySwarms - Open Source Software Contribution

To learn more about collaborative development with version control systems as well as contributing to open source projects on GitHub, I made two contributions to the PySwarms project. PySwarms is a Python module for particle swarm optimization research.

Toxic Comment Classification - Natural Language Processing

I performed a detailed analysis of Wikipedia comments and built a model that classifies them as toxic or nontoxic. The final model is a support vector machine that uses a Naive Bayes feature weight transformer to improve performance. The data was obtained from Kaggle.

Galaxy Zoo Challenge - Image Classification with PyTorch

This is an image classification project that I completed for my independent study at Old Dominion University. The dataset was obtained from Kaggle’s Galaxy Zoo Challenge. I chose astronomy datasets for my independent study because I enjoy learning about the topic and they presented an opportunity to sharpen my deep learning and PyTorch skills.

ODUConnect - Software Engineering Group Project

ODUConnect is a web application built on a MSSQL Database / Flask API / ReactJS Front End stack. I let the front end and API development and built the continuous deployment, among other responsibilities.

PLAsTiCC Astronomical Classification - Light Curve Sequence Analysis

This is a sequence analysis project that I completed for my independent study at Old Dominion University. I used the open source dataset from Kaggle’s PLAsTiCC Astronomical Classification Challenge. In the project I analyzed light curves, engineered over a hundred metadata features, and built a LightGBM model to predict the object classes.

publications

talks

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.