The objective of this project is to build: i.) a web app returning the percentage likelihood of a boxing fight ending in a particular outcome ii.) an interactive dashboard with a custom boxer rank by division, comparisons between different fight stances and age groups based on the data extracted
High Level Blog Post: https://hackernoon.com/d-nr1o32po (explaining data extraction, cleanup and building the interactive dashboard) https://hackernoon.com/i-built-a-boxing-prediction-web-app-on-shiny-here-is-how-jz8932xt (explaining the process of enriching my data with punch stats scraped from CompuBox, building the machine learning model and the shiny web app)
Heroku Visualization App: https://boxingvisualisationdashboard.herokuapp.com/
As mentioned this project also includes the code relevant to my boxing prediction web app.
Under the ml model folder I included the code relevant to building the model used for this project (randomforestmodel.ipynb) Along with the code specifically relevant for building the shiny web app interface (under the folder app/boxing).
High Level Blog Post: https://hackernoon.com/i-built-a-boxing-prediction-web-app-on-shiny-here-is-how-jz8932xt Web App: https://thebeyonder.shinyapps.io/boxingapp2/ (please refresh brower in the unlikely event that you encounter an error message)