Google Cloud

Google Cloud Certified – Professional Machine Learning Engineer

PR000269
Trending

Validate your skills as a Professional Machine Learning Engineer with exam code PR000269 in Google Cloud.

495 questions 0 views Free
Start Mock Test Timed · Full-length · Scored

Questions 491–495 of 495

Q491

A company needs to train a machine learning model using its existing data in CSV format stored in Google Cloud Storage. What should they do first?

  • A Upload data to BigQuery
  • B Convert CSV to JSON format
  • C Use Dataflow to transform data
  • D Use AI Platform's Training Services
Explanation AI Platform can directly use CSV data for training without conversion.
Q492

What happens when you set the 'min_instances' parameter to zero in a Google Cloud Machine Learning model?

  • A Model scales instantly
  • B No instances are allocated
  • C Model runs continuously
  • D Instances are cached for performance
Explanation Setting 'min_instances' to zero allocates no active instances, saving costs when idle.
Q493

Which Google Cloud service is best for orchestration?

  • A Cloud Functions
  • B Cloud Run
  • C Cloud Composer
  • D Cloud Storage
Explanation Cloud Composer is designed for workflow orchestration, while the others serve different purposes.
Q494

A retail company is analyzing user purchase patterns using real-time data. What Google Cloud service should they predominantly use?

  • A BigQuery
  • B Pub/Sub
  • C Dataproc
  • D Firestore
Explanation Pub/Sub is ideal for real-time messaging, whereas the others are for different data analysis purposes.
Q495

You are configuring a machine learning model with a skewed dataset. What should you do to improve model performance?

  • A Use all features regardless of importance
  • B Oversample the minority class
  • C Ignore preprocessing steps
  • D Use default hyperparameters
Explanation Oversampling the minority class helps balance the dataset, enhancing model performance compared to the other options.