Using Machine Learning to Improve Baseball Swings

In recent years, there has been growing interest in the use of neural networks to analyze and improve athletic performance. One area where this technology has been applied is in baseball, specifically in analyzing the mechanics of a player's swing. Traditionally, coaches and scouts have relied on subjective observations to evaluate a player's swing, but with the use of neural networks and other advanced technologies, it is possible to gather objective data and make more informed decisions about training and development. During a swing, the foot pressure is particularly important. It provides insight into the transfer of weight from the back foot to the front foot, which is a critical aspect of a powerful and effective swing.  We can then use neural networks to identify patterns and make predictions about a player's performance by collecting data on foot pressure during a swing. We believe that the application of neural networks to baseball swing analysis has the potential to improve the way coaches and scouts evaluate players and help them improve their performance. In this work, we seek to apply machine learning techniques, including recurrent neural networks and convolutional neural networks, to analyze baseball swings with data collected from foot pressure sensors.