Question 2

Explanation

Correct Answer:

✅ All of the above

Detailed Explanation of the Correct Answer:

✅ "All of the above"

• Databricks AutoML automatically calculates multiple evaluation metrics for regression problems, including:

  1. Mean Absolute Error (MAE)

  2. Coefficient of Determination (R-squared, R²)

  3. Root Mean Square Error (RMSE)

  4. Mean Square Error (MSE)

Breakdown of Each Regression Metric:

MetricDescriptionBest Use CaseMean Absolute Error (MAE)Measures the average absolute differences between actual and predicted values.Useful when all prediction errors should be equally weighted.R-squared (R², Coefficient of Determination)Represents the proportion of variance explained by the model.Measures goodness of fit—higher values mean better fit.Root Mean Square Error (RMSE)Measures the square root of the average squared errors.Penalizes large errors more heavily than MAE.Mean Square Error (MSE)Measures the average squared errors between actual and predicted values.Used when large errors should be penalized more significantly.

Example: Evaluating a Regression Model in AutoML

from pyspark.ml.evaluation import RegressionEvaluator
 
# Initialize evaluator
evaluator = RegressionEvaluator(predictionCol="prediction", labelCol="actual")
 
# Compute different metrics
mae = evaluator.setMetricName("mae").evaluate(preds_df)
r2 = evaluator.setMetricName("r2").evaluate(preds_df)
rmse = evaluator.setMetricName("rmse").evaluate(preds_df)
mse = evaluator.setMetricName("mse").evaluate(preds_df)
 
print(f"MAE: {mae}, R-squared: {r2}, RMSE: {rmse}, MSE: {mse}")

• Databricks AutoML logs these metrics for every model run, enabling easy comparison.

Why Other Options Are Incorrect:

❌ "Mean Absolute Error (MAE)"

• Incorrect, because MAE is just one of several evaluation metrics used in regression.

❌ "Coefficient of Determination (R-squared)"

• Incorrect, because AutoML also computes MSE, RMSE, and MAE.

❌ "Root Mean Square Error (RMSE)"

• Incorrect, because other regression metrics are calculated along with RMSE.

❌ "Mean Square Error (MSE)"

• Incorrect, because AutoML tracks additional metrics like MAE and R².

Final Conclusion:

In an AutoML regression experiment, the automatically calculated evaluation metrics include MAE, R², RMSE, and MSE, so the correct answer is:
✅ "All of the above"