This is a "just-for-fun" datascience project with a goal of revolutionizing the NFL draft in the coming years. The project contains a dataset of NFL WRs drafted from 2018 to 2020, with their measureables and college statistics. Their wAv/year is then used to train a neural network to predict how valuable new WRs entering the league is going to be. All code is in python, and the data is in xlsx files.
- Data-folder: This folder contains the data used in training and testing the model.
- Regressor: Contains reading the data, plotting PLC, the model, plotting the loss over the epochs as well as training and testing the model.
- SpreadFiller: Fills the missing values in the data with averages of the other values in the column.
- SpreadFixer: Converts the measureables of the athletes from ft to cm, and from lbs to kg.
- NLP: Not used in the project, only for experimenting with adding some NLP features to the model in the future.
- More information in the dataset:
- Rating the college
- Information about position (slot, posession, deep threat, etc)
- Personality (maturity issues, etc)
- Injuries
- More draft classes in the dataset (2021, 2022 missing)
- In the future, not only WRs.
python3 regressor.pyGitHub LinkedIn Email: Vic@Norris.no