This is a Machine Learning test using 2019 NCAA March Madness data and https://github.com/KINGdotNET/March-Madness-ML project originally initiated by github user (adeshpande3).
Tools that I’ve used to run my Machine Learning projects.
- Anaconda, and install dependencies using Anaconda command line as administrator.
- conda create -n tensorflow python=3.6 // This is the current version running in my environment, so change it with your version.
- conda activate tensorflow // a simple test is to run “import tensorflow as tf” in your python command line.
- conda install -c anaconda scikit-learn
- conda install -c anaconda pandas
- conda install -c anaconda keras
- conda install -c anaconda matplotlib
- conda install -c anaconda py-xgboost
After installing all the dependencies, run “DataPreProccessing.py” then after that run “python MarchMadness.py” to see the results. Here’s my 2019 NCAA March Madness selection through the use of Machine Learning.
Running the DataPreProcessing from Year 2000 to current year, as shown below.
First run of MarchMadness.py.
and the result.
The Machine Learning code predicted Duke will win all the way against Virginia. My personal assessment is Duke will win against North Carolina. We will find out very soon.