The goal is to predict ‘TOTAL_GROSS’ earnings using an XGBoost regression model.

Data Preparation:

Features: [‘RETRO’, ‘DETAIL’, ‘OVERTIME’, ‘INJURED’, ‘QUINN_EDUCATION’, ‘REGULAR’]

Target: ‘TOTAL_GROSS’

Top Departments Selection:

The top 10 departments by employee count are selected for analysis.

Train-Test Split:

The data is split into training and testing sets (80% training, 20% testing).

XGBoost Regression Model:

An XGBoost Regressor is created and trained on the training set.

Prediction:

The model predicts ‘TOTAL_GROSS’ on the test set (‘X_test’).

Evaluation:

 

The R2 score is calculated to assess the model’s performance on the test data.

Output:

 

The R2 score for the XGBoost model is 0.3542, indicating the model explains 35.42% of the variance in ‘TOTAL_GROSS’.