Databricks Machine Learning Associate Question 603
Select 3You have trained a regression model to predict house prices using a dataset containing features such as the number of bedrooms, square footage, and location. After evaluating the model, you obtain the following metrics: RMSE = 25,000, MAE = 20,000, and R-squared = 0.85. Which of the following conclusions can you draw from these metrics?
- A
The RMSE being higher than the MAE indicates that the model's predictions include some larger errors.
- B
An R-squared value of 0.85 suggests that 85% of the variance in the target variable is explained by the model.
- C
A lower MAE compared to RMSE always means that the model is underfitting.
- D
If R-squared is close to 1, the model perfectly predicts all data points in the test set.
- E
The combination of RMSE, MAE, and R-squared metrics provides insights into both the accuracy of the model and the quality of the predictions.
Show answer and explanation
Correct answers: A, B, E
Explanation
Evaluating a regression model requires understanding various metrics such as RMSE, MAE, and R-squared. RMSE and MAE provide insights into the magnitude and average of prediction errors, while R-squared measures how well the model explains the variance in the data. Combining these metrics helps assess both the accuracy and quality of the model's predictions. In this scenario, the provided metrics indicate a reasonably accurate model with some larger prediction errors, as shown by the RMSE being higher than the MAE.
- A. Correct.
Correct. When RMSE is higher than MAE, it indicates that the model's prediction errors include some large deviations, as RMSE penalizes larger errors more heavily compared to MAE.
- B. Correct.
Correct. R-squared measures how well the model explains the variance in the target variable, and a value of 0.85 indicates that 85% of the variance is explained by the model.
- C. Incorrect.
Incorrect. A lower MAE compared to RMSE does not necessarily indicate underfitting. It simply reflects the distribution of errors, with RMSE being more sensitive to large errors.
- D. Incorrect.
Incorrect. An R-squared value close to 1 indicates high explanatory power but does not guarantee perfect predictions for all data points. Other metrics, like RMSE or MAE, are needed to assess prediction accuracy.
- E. Correct.
Correct. RMSE evaluates the magnitude of errors, MAE measures the average error, and R-squared evaluates variance explained, providing a comprehensive view of model performance.