Context

HomePricePredictor is a machine learning model designed to predict housing prices based on various features such as location, size, and amenities. After initial deployment, the model showed inconsistent performance across different datasets, raising concerns about its reliability. The data science team implemented k-fold cross-validation to assess the model's generalization capabilities better.

Current Performance

The Problem

While cross-validation improved the model's RMSE and R² significantly, the increased training time raises questions about efficiency. The team needs to understand the trade-offs between model performance and computational cost.

Requirements

1. Analyze the effect of cross-validation on model performance metrics.
2. Identify potential causes for the training time increase.
3. Recommend strategies to optimize training time without sacrificing performance.
4. Discuss the importance of cross-validation in preventing overfitting.

Constraints

• The model must be retrained weekly due to new data influx, limiting the maximum acceptable training time to 2.5 hours.