Databricks Machine Learning Associate exam dumps

Databricks Machine Learning Associate practice question 263 of 656

Databricks Certified Machine Learning Associate. Associate level, Databricks. Free question with the correct answer and a full explanation.

Databricks Machine Learning Associate Question 263

Single answer

A data scientist is building a regression model to predict house prices. Since the price distribution is highly skewed, they apply a logarithmic transformation to the target variable (house price) before training the model. After evaluating the trained model, the Root Mean Squared Error (RMSE) is reported as 0.35. However, the data scientist forgets to exponentiate the RMSE to match the original scale of the target variable. What is the consequence of this mistake?

  1. A

    The RMSE value is correctly reported and does not require any further adjustments.

  2. B

    The RMSE is underestimated and does not represent the error on the original scale of the target variable.

  3. C

    The RMSE will be equivalent to the Mean Squared Error (MSE) on the transformed scale.

  4. D

    Exponentiating the RMSE would significantly reduce its value on the transformed scale.

Show answer and explanation

Correct answer: B

Explanation

When the target variable is logarithmically transformed, the RMSE is computed on the transformed scale. To interpret the RMSE on the original scale of the target variable, it needs to be exponentiated. Failing to do so results in an underestimated error when interpreting the RMSE in the context of the original scale. This is a common mistake when working with log-transformed targets and must be addressed to ensure proper evaluation of model performance.

  • A. Incorrect.

    This is incorrect because the RMSE computed on the log-transformed scale needs to be exponentiated to reflect the error on the original scale of the target variable.

  • B. Correct.

    This is correct because the RMSE on the log-transformed scale does not directly represent the error on the original scale of the target variable. Failing to exponentiate the RMSE results in an underestimated value when interpreting it in the context of the original scale.

  • C. Incorrect.

    This is incorrect because RMSE and MSE are different metrics. RMSE is the square root of MSE, so they cannot be equivalent.

  • D. Incorrect.

    This is incorrect because exponentiating the RMSE would increase its value (not reduce it) on the original scale of the target variable, as it would account for the exponential nature of the log transformation.

Timed practice exam

Take a Databricks Machine Learning Associate practice test under exam conditions

48 questions in 90 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam