Exam NCA-AIIO Topic 1 Question 107 Discussion
Actual exam question for NVIDIA's NCA-AIIO exam
Question #: 107
Topic #: 1
Question #: 107
Topic #: 1
You have developed two different machine learning models to predict house prices based on various features like location, size, and number of bedrooms. Model A uses a linear regression approach, while Model B uses a random forest algorithm. You need to compare the performance of these models to determine which one is better for deployment. Which two statistical performance metrics would be most appropriate to compare the accuracy and reliability of these models? (Select two)
Suggested Answer: C,E Vote an answer
For regression tasks like predicting house prices (a continuous variable), the appropriate metrics focus on accuracy and reliability of numerical predictions:
* Mean Absolute Error (MAE)(C) measures the average absolute difference between predicted and actual values, providing a straightforward indicator of prediction accuracy. It's intuitive and effective for comparing regression models.
* R-squared (Coefficient of Determination)(E) indicates how well the model explains the variance in the target variable (house prices). A higher R-squared (closer to 1) suggests better fit and reliability, making it ideal for comparing Model A (linear regression) and Model B (random forest).
* F1 Score(A) is used for classification tasks, not regression, as it balances precision and recall.
* Learning Rate(B) is a hyperparameter for training, not a performance metric.
* Cross-Entropy Loss(D) is typically used for classification, not regression tasks like this.
MAE (C) and R-squared (E) are standard metrics in NVIDIA RAPIDS cuML and other ML frameworks for regression evaluation.
* Mean Absolute Error (MAE)(C) measures the average absolute difference between predicted and actual values, providing a straightforward indicator of prediction accuracy. It's intuitive and effective for comparing regression models.
* R-squared (Coefficient of Determination)(E) indicates how well the model explains the variance in the target variable (house prices). A higher R-squared (closer to 1) suggests better fit and reliability, making it ideal for comparing Model A (linear regression) and Model B (random forest).
* F1 Score(A) is used for classification tasks, not regression, as it balances precision and recall.
* Learning Rate(B) is a hyperparameter for training, not a performance metric.
* Cross-Entropy Loss(D) is typically used for classification, not regression tasks like this.
MAE (C) and R-squared (E) are standard metrics in NVIDIA RAPIDS cuML and other ML frameworks for regression evaluation.
by Hiram at Jun 20, 2026, 06:59 AM
0
0
0
10
Comments
Upvoting a comment with a selected answer will also increase the vote count towards that answer by one. So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.
Report Comment
Commenting
You can sign-up / login (it's free).