DARE-Newcrest Data Challenge: Flotation Plant Data

This repository contains digital resources for the DARE-Newcrest data challenge.

Data

The data are available in the data folder. We will be using the files:

- data/hourly_train_data.csv
- data/hourly_test_data.csv

Code

The data were prepared using the notebook 00_Data_Preprocessing.ipynb.

You will use the notebook 01_Data_Challenge.ipynb to complete the challenge.

This has code for loading the data, splitting into train/validation/test, scaling the data, and an example training a simple linear regression model and neural network.

Challenge

Your challenge is to develop a better model for prediction.

Model performance will be assessed using Mean Absolute Error, Mean Squared Error and Coefficient of Determination (R²) metrics.

N.B. The notebook contains some standard functions for evaluating and plotting model performance.