Answer-first summary for fast verification
Answer: Array transformations, which involve operations on complex data types like lists within the CSV file.
AutoML supports transformations for Text, Number, Timestamp, and Categorical data types when the data is in CSV format. However, for complex data types like Arrays and Structs, transformations are only available if the data is in BigQuery, which natively supports these types. This limitation is due to the CSV format's inability to natively represent complex data structures, unlike BigQuery. For more details, refer to the official documentation: [Vertex AI Data Types for Tabular Data](https://cloud.google.com/vertex-ai/docs/datasets/data-types-tabular) and [Compound Data Types](https://cloud.google.com/vertex-ai/docs/datasets/data-types-tabular#compound_data_types).
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
Your team is developing a multiclass logistic regression model using tabular data in Vertex AI with AutoML. The data is stored in a CSV file in Cloud Storage. AutoML can perform various transformations on the data to optimize its use for the model. Given the constraints of using CSV format in Cloud Storage and the need to ensure the model's scalability and performance, which of the following transformation types is not permitted? Choose the one correct option.
A
Text transformations, including tokenization and n-gram generation, to process textual data within the CSV.
B
Number transformations, such as normalization and scaling, to adjust numerical data for better model performance.
C
Timestamp transformations, including parsing and formatting, to handle date and time data effectively.
D
Array transformations, which involve operations on complex data types like lists within the CSV file.
E
Categorical transformations, including one-hot encoding and label encoding, to manage categorical variables.