Transforming and standardizing features: Practice Questions — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)

Practice Questions: Transforming and Standardizing Features

This set of multiple-choice questions is designed to help you prepare for the Transforming and Standardizing Features section of the NVIDIA-Certified Professional: Accelerated Data Science exam. Each question focuses on key concepts related to feature transformation and standardization using GPU-accelerated libraries like cuDF and RAPIDS.

1. Which of the following methods in cuDF is typically used to standardize a numerical feature by removing the mean and scaling to unit variance?

   • A. cudf.Series.min()
   • B. cudf.Series.mean()
   • C. cudf.Series.std()
   • D. Manual calculation using mean() and std() to apply z-score scaling

   Correct answer: D

   Explanation: cuDF does not provide a direct standardization method; you typically compute the mean and standard deviation manually and apply z-score scaling: (x - mean) / std.

2. When transforming categorical features for machine learning models using RAPIDS cuDF, which technique is commonly used to convert categories into numerical values without implying ordinal relationships?

   • A. Label encoding
   • B. One-hot encoding
   • C. Min-max scaling
   • D. Standard scaling

   Correct answer: B

   Explanation: One-hot encoding creates binary columns for each category, avoiding ordinal assumptions. Label encoding assigns integer values but can imply order, which may mislead models.

3. Which cuDF function helps to efficiently apply a custom transformation to a column of data during feature engineering?

   • A. apply_rows()
   • B. groupby()
   • C. merge()
   • D. fillna()

   Correct answer: A

   Explanation: apply_rows() allows you to apply a custom CUDA kernel row-wise, enabling efficient feature transformations on GPU.

4. Why is feature standardization important before training many machine learning models?

   • A. It reduces the number of features
   • B. It ensures all features have the same scale, improving model convergence
   • C. It converts categorical variables into numerical ones
   • D. It removes missing values from the dataset

   Correct answer: B

   Explanation: Standardizing features to have zero mean and unit variance helps gradient-based models converge faster and prevents features with larger scales from dominating.

5. In RAPIDS cuDF, which method can be used to normalize a feature to a fixed range, such as [0,1]?

   • A. Series.min() and Series.max() with manual scaling
   • B. Series.mean()
   • C. Series.std()
   • D. Series.unique()

   Correct answer: A

   Explanation: Min-max normalization requires computing the minimum and maximum values and scaling each value accordingly: (x - min) / (max - min).

6. Which of the following is a key advantage of using RAPIDS libraries like cuDF for feature transformation and standardization?

   • A. Automatic feature selection
   • B. GPU acceleration for faster data processing
   • C. Built-in model training
   • D. Cloud-based data storage

   Correct answer: B

   Explanation: RAPIDS cuDF leverages GPU acceleration to perform data transformations and standardization much faster than CPU-based libraries like pandas.

7. When standardizing a feature with outliers using cuDF, which approach can help reduce the influence of extreme values?

   • A. Use z-score standardization directly
   • B. Apply robust scaling techniques such as median and interquartile range
   • C. Ignore outliers
   • D. Use one-hot encoding

   Correct answer: B

   Explanation: Robust scaling uses statistics less sensitive to outliers (median, IQR) to standardize features, improving model robustness.

8. Which RAPIDS tool can be used to generate synthetic data for augmenting feature sets during data preparation?

   • A. cuDF
   • B. cuML
   • C. cuGraph
   • D. cuSignal

   Correct answer: B

   Explanation: cuML includes utilities for synthetic data generation, useful for augmenting datasets during feature engineering and transformation.

More in this topic

Transforming and standardizing features: Quick Reference — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Data cleansing and preprocessing with cuDF and pandas: Practice Questions — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Data cleansing and preprocessing with cuDF and pandas: Common Mistakes — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Generating synthetic data with cuDF and RAPIDS: Practice Questions — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Monitoring pipeline bottlenecks: Worked Example — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Monitoring pipeline bottlenecks: Practice Questions — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Generating synthetic data with cuDF and RAPIDS: Quick Reference — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Generating synthetic data with cuDF and RAPIDS: Worked Example — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Monitoring pipeline bottlenecks — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Monitoring pipeline bottlenecks: Quick Reference — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Monitoring pipeline bottlenecks: Common Mistakes — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Generating synthetic data with cuDF and RAPIDS: Common Mistakes — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Transforming and standardizing features: Common Mistakes — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Data Preparation — NVIDIA-Certified Professional: Accelerated Data ScienceData cleansing and preprocessing with cuDF and pandas — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Data cleansing and preprocessing with cuDF and pandas: Quick Reference — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Generating synthetic data with cuDF and RAPIDS — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Transforming and standardizing features: Worked Example — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Data cleansing and preprocessing with cuDF and pandas: Worked Example — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)Transforming and standardizing features — Data Preparation (NVIDIA-Certified Professional: Accelerated Data Science)