Preparing datasets for machine learning: Practice Questions — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)

Practice Questions: Preparing Datasets for Machine Learning

These multiple-choice questions are designed to help you prepare for the Preparing Datasets for Machine Learning section of the NVIDIA-Certified Associate: Generative AI LLM exam, focusing on data preprocessing, feature engineering, and dataset readiness for training large language models.

1. Which of the following is the primary purpose of feature scaling in dataset preparation?

   • A. To reduce the number of features in the dataset
   • B. To normalize feature values to a common scale for better model convergence
   • C. To remove missing values from the dataset
   • D. To increase the size of the dataset

   Correct Answer: B

   Explanation: Feature scaling normalizes feature values so that they are on a similar scale, which helps many machine learning algorithms converge faster and perform better.

2. When preparing text data for training a large language model, which preprocessing step is essential?

   • A. One-hot encoding of numerical features
   • B. Tokenization of text into smaller units such as words or subwords
   • C. Removing all punctuation marks without exception
   • D. Converting text to binary format

   Correct Answer: B

   Explanation: Tokenization breaks text into units that the model can process, which is a fundamental step in preparing text data for language models.

3. Which GPU-accelerated library is commonly used for efficient data manipulation and preprocessing in Python?

   • A. NumPy
   • B. Pandas
   • C. cuDF
   • D. Matplotlib

   Correct Answer: C

   Explanation: cuDF is a GPU-accelerated dataframe library that speeds up data manipulation tasks, making it suitable for large datasets in AI workflows.

4. What is the main reason for splitting a dataset into training, validation, and test sets?

   • A. To increase the dataset size
   • B. To evaluate model performance on unseen data and prevent overfitting
   • C. To reduce the number of features
   • D. To speed up data preprocessing

   Correct Answer: B

   Explanation: Splitting data helps assess how well the model generalizes to new data and prevents overfitting by tuning on validation data.

5. Which technique is used to handle missing data in a dataset before training?

   • A. Data augmentation
   • B. Imputation
   • C. Feature scaling
   • D. Tokenization

   Correct Answer: B

   Explanation: Imputation fills in missing values using statistical methods (mean, median, mode) or model-based approaches to prepare the dataset for training.

6. Why is one-hot encoding applied to categorical features in dataset preparation?

   • A. To reduce dataset size
   • B. To convert categorical data into a numerical format understandable by machine learning models
   • C. To normalize numerical features
   • D. To remove irrelevant features

   Correct Answer: B

   Explanation: One-hot encoding transforms categorical variables into binary vectors, enabling models to interpret categorical data correctly.

7. Which of the following best describes the role of data augmentation in dataset preparation?

   • A. Removing outliers from the dataset
   • B. Increasing dataset diversity by creating modified versions of existing data
   • C. Converting categorical data into numerical data
   • D. Splitting the dataset into training and test sets

   Correct Answer: B

   Explanation: Data augmentation artificially expands the dataset by generating variations, which helps improve model robustness, especially in image and text data.

8. What is the advantage of using GPU-accelerated data preprocessing when preparing datasets for machine learning?

   • A. It reduces the need for data cleaning
   • B. It speeds up data manipulation and feature engineering on large datasets
   • C. It automatically selects the best features
   • D. It eliminates the need for splitting datasets

   Correct Answer: B

   Explanation: GPU acceleration significantly reduces the time required for data preprocessing tasks, enabling faster iteration and model training.

More in this topic

Preparing datasets for machine learning: Worked Example — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Data analysis and visualization: Quick Reference — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)GPU-accelerated data manipulation: Worked Example — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)GPU-accelerated data manipulation — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Data analysis and visualization: Worked Example — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Data analysis and visualization — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Preparing datasets for machine learning — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)GPU-accelerated data manipulation: Practice Questions — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Preparing datasets for machine learning: Quick Reference — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)GPU-accelerated data manipulation: Quick Reference — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Data analysis and visualization: Common Mistakes — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Preparing datasets for machine learning: Common Mistakes — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Data Analysis and Visualization — NVIDIA-Certified Associate: Generative AI LLMData preprocessing and feature engineering — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)GPU-accelerated data manipulation: Common Mistakes — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)Data analysis and visualization: Practice Questions — Data Analysis and Visualization (NVIDIA-Certified Associate: Generative AI LLM)