Answer-first summary for fast verification
Answer: Utilize BigQuery ML to execute several regression models within BigQuery, facilitating rapid experimentation and performance analysis using SQL queries., Combine the use of BigQuery ML for initial rapid experimentation with Vertex AI for more complex model training and deployment, ensuring a comprehensive approach to model development and evaluation.
BigQuery ML (Option C) is the most suitable choice for the team's primary need for a self-service, efficient tool that allows quick experimentation with various models directly within BigQuery, using SQL queries. It simplifies the process for users with varying levels of expertise by providing automated hyperparameter tuning and comprehensive evaluation metrics. However, for scenarios requiring more complex model training beyond what BigQuery ML offers, combining it with Vertex AI (Option E) provides a comprehensive solution that covers both rapid experimentation and advanced model development needs.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
A company aims to leverage its marketing data stored in BigQuery to predict sales outcomes efficiently. The data science team, with varying levels of expertise, seeks a self-service solution to quickly experiment with and evaluate multiple machine learning models, from simple linear regressions to complex neural networks, without extensive setup or deep data science knowledge. The solution should also provide comprehensive evaluation metrics and support automated hyperparameter tuning to streamline the model selection process. Considering these requirements, which Google Cloud tool(s) would best meet the team's needs? (Choose one correct option unless option E is present, then choose two.)
A
Develop and train custom TensorFlow models using Vertex AI, directly accessing data from BigQuery, to leverage a wide range of ML algorithms and neural network architectures.
B
Employ Vertex AI Workbench user-managed notebooks to write and execute scikit-learn code for various ML algorithms, enabling detailed performance analysis and customization.
C
Utilize BigQuery ML to execute several regression models within BigQuery, facilitating rapid experimentation and performance analysis using SQL queries.
D
Process the data in BigQuery with Dataproc and implement various machine learning models using SparkML for distributed computing capabilities.
E
Combine the use of BigQuery ML for initial rapid experimentation with Vertex AI for more complex model training and deployment, ensuring a comprehensive approach to model development and evaluation.