DARE - Deluxe Data Challenge I

This repository contains the data and example code for the DARE Deluxe Data Challenge I.

Data

Data were prepared using the Jupyter notebook 00_Data_Preprocessing.ipynb. We will be using the files:

data/daily_train_X_data.csv
data/daily_train_y_data.csv
data/daily_test_X_data.csv
data/daily_test_y_data.csv

Code

You should use either...

• Python Jupyter notebook 01_Data_Challenge.ipynb or
• R Markdown file R_01_Data_Challenge.Rmd

to complete the challenge.

Those files have code for...

• Loading data.
• Scaling data.
• Splitting data into train/validation/test partitions.
• Example model training via simple linear regression and a neural network.
• Evaluating and plotting model performance.

Your task: develop a better model for prediction.

Predictive model performance will assessed with BSS, MAE, MSE and R² metrics.

Glossary

Acronym	Definition
BSS	Brier Skill Score
MAE	Mean Absolute Error
MSE	Mean Squared Error
R2	Coefficient of Determination

Acknowledgements

Contributors: Joshua Simmons and Travis Stenborg.