Capstone Project - Springboard Machine Learning Bootcamp, 2020

App deployed on Heroku

Predicting an NBA's team Points per Game based on other features

Author: Ben Basuni

Key Question: Can we use historic NBA data and machine learning to predict a team's total points for that game?

Project Overview

• Deployment: I've built a very simple machine learning model that demonstrates what I've learned in this bootcamp. It's a very simple Machine Learning model that

• Background: NBA games produce a wide array of statistical data to be analyzed.
  We will be using this data to take a statistical viewpoint of an NBA team and how many points a team will make.

• Datasets & Input: https://www.kaggle.com/drgilermo/nba-players-stats

• Solution Statement: The problem to predict the total points in a game given other features is a regression problem. We will take existing data and input, and do some data manipulation and see if we can come up with statistical data that . This mimicks an NBA Sports Analysis that heavily uses statistics to cast his vote.

• Benchmark Model: Default SciKit-Learn Logistic Regression, RandomForestRegressor, and a Neural Network will be used as a benchmark/baseline. Several models will then be explored to improve over the benchmark including other ensemble and tree-based models, Support-Vector Machines (SVM), XGBoost.

• Evaluation Metrics: R-squared and RMSE will be used for the regression part of the model

Project Folders

• assets - neural networking models
• nba-dataset - csv files that are used
• static - Images/CSS/JS files
• templates - flask html files
• ipynb - Jupyter notebooks that I worked on in Kaggle to help support my project

How to Run

pip3 install -r requirements.txt