A data scientist needs to create a model for predictive maintenance. The model will be based on historical data to identify rare anomalies in the data. The historical data is stored in an Amazon S3…

Question

Options

• AImport the data by using the None option.
• BImport the data by using the Stratified option.
• CImport the data by using the First K option. Infer the value of K from domain knowledge.
• DImport the data by using the Randomized option. Infer the random size from domain knowledge.

• A
  9% (3)
• B
  6% (2)
• C
  82% (27)
• D
  3% (1)

Explanation

The First K option in Amazon SageMaker Data Wrangler allows you to sample the first K records of the dataset. This is efficient when you have large datasets and want to perform exploratory data analysis (EDA) with a smaller subset, thereby reducing the need for significant compute resources. By using domain knowledge to infer the appropriate value of K, the data scientist can ensure that a representative subset of the data is used for EDA, minimizing the required compute None option would import the entire dataset, potentially requiring a lot of compute resources, which goes against the requirement of minimizing resource usage. Stratified option would sample data based on the distribution of a specific feature, but this could still require more compute than necessary, especially if the full dataset is large. Randomized option introduces randomness, but it may not be as efficient as selecting a known representative subset (First K) based on domain

No community discussion yet for this question.