These are a series of projects I have completed in an effort to improve my programming skills and work with data that I find interesting! Within each folder you can find a Jupyter Notebook that walks through the code and my thought process going through each project. By no means are they perfect, but I hope that as I continue to complete these passion projects the quality of every subsequent one will improve :)
1. Relationship between Bat Speed and High Performance Metrics
This was a small project done as an opportunity to practice my coding skills as well as to begin to dive into the newly released open-source High Performance dataset provided by Driveline Baseball. The goal of this project was to try and begin to understand the relationship between select high performance metrics and bat speed.
2. Investigating the Kinematics and Kinetics of Baseball Hitting
This was another project done as an opportunity to practice my programming skills as well as to try my hand at using statistical parametric mapping to understand the open-source baseball hitting dataset provided by Driveline Baseball. The goal of this project was to investigate the differences in the kinematics and kinetics between 'fast', 'average', and 'slow' exit velocity groups.
3. Exploring the Use of Raw Force Time Data Points to Evaluate Jump Performance
This code was created for the OBEL Lab 'Hack-a-thon' where we were tasked to use raw force plate data to evaluate CMJ performance of various UWaterloo athletes. The goals were to:
- Be able to demarcate each phase of the jump based on F/T data provided
- Calculate specific output metrics
- Determine whether there are any variables that have strong relationships with jump height
- Predict the draft round for half of the athletes
4. A Flexible EMG Processing Pipeline to Visualize Injury Risk
This program was created with a few goals in mind:
- Write a code that can process raw EMG according to best practices and visualize the final results in the form of amplitude probability distribution functions.
- Create a pipeline that can be flexible and handle multiple muscles and trials.
- Be as user-friendly as possible so that someone without formal training could run the code and get injury risk data.
To see the program in action, please see the Jupyter Notebook!