Projects

Computational Study of Neural Action Potentials Using the Hodgkin-Huxley Model

Investigation and computational study of the Hodgkin-Huxley model of the action potential in neurons. In this project, we explore both steady-state behavior and dynamic responses to various stimuli.

Tech Stack: Python, Scipy, Matplotlib, Git, Jupyter Notebooks

Decoding Baseball Pitcher Intentions through Recurrent Neural Networks

Development of RNN to predict baseball pitches. Model receives pitcher video data, makes a future pitch prediction, then demonstrates how it comes to that decision with guided backpropogation and saliency maps.

Tech Stack: Python, PyTorch, Torchvision, Pandas, Scikit-learn, Seaborn, Git

Introducing MyTorch: A Fully Custom, Tailored Deep Learning Framework

Object-oriented deep learning library called "MyTorch". The library has the ability to call functions similarly to the PyTorch library. MyTorch has working loss functions, optimizers (stochastic gradient descent), and forward/backward passes.

Tech Stack: Python, NumPy

Utilizing Convolutional Neural Networks for Human Eye Ultrasonography Analysis

Developed multi-output CNN to analyze human eye ultrasound images. Goal: to create a model that can accurately determine the vertical and horizontal diameter of the human eye.

Tech Stack: Python, PyTorch, Pandas, Scikit-learn, Numpy, Seaborn, Git, Unix

Forecasting Geothermal Power Plant Feasibility with Neural Networks

Developed NN model to predict geothermal heatflow residuals. The classification model uses a subset of the 27 provided features to classify the residuals into four classes.

Tech Stack: Python, PyTorch, Pandas, Torchvision, Scikit-learn, Numpy, Seaborn, Git, Unix

Assessing Geothermal Power Plant Suitability with Convolutional Neural Networks

Developed CNN model to detect geothermal favorability locations in the western US utilizing detrended elevation maps.

Tech Stack: Python, PyTorch, Pandas, Torchvision, Scikit-learn, Numpy, Seaborn, Git, Unix

Roped Together, a 3D Physics Based Mountain Climbing Game

Physics based mountain climbing game with over 300 players. Winner of the regional Oregon Game Project Challenge.

Tech Stack: Unity, C#, Git, GitKraken, Trello, Blender