Google Professional Machine Learning Engineer Question 75
Select 2Google Cloud PlatformYou are tasked with building a machine learning model to predict customer churn for a subscription-based service. The dataset is stored in BigQuery and includes both numerical and categorical features. You decide to use AutoML Tables in Google Cloud to simplify the process. Which steps should you take to ensure the best model performance?
- A
Preprocess your dataset by ensuring all categorical features are converted to one-hot encoded values before uploading to AutoML Tables.
- B
Split your dataset into training, validation, and test sets before importing it into AutoML Tables.
- C
Provide a properly formatted schema file with column data types and labels when importing the dataset into AutoML Tables.
- D
Enable explainable AI during the model training process to understand feature importance.
- E
Ensure that your dataset includes a clear target column and has no missing values before importing it into AutoML Tables.
Show answer and explanation
Correct answers: D, E
Explanation
AutoML Tables simplifies the machine learning process by automating many tasks, such as preprocessing, data splitting, and schema inference. However, enabling explainable AI can provide valuable insights into the model's decision-making process, and ensuring clean data with a valid target column is critical for achieving optimal model performance.
- A. Incorrect.
Incorrect. AutoML Tables automatically handles preprocessing, including encoding of categorical features, so manual one-hot encoding is unnecessary.
- B. Incorrect.
Incorrect. AutoML Tables automatically splits your dataset into training, validation, and test sets, so you do not need to perform this step manually.
- C. Incorrect.
Incorrect. AutoML Tables infers column data types and relationships automatically, so providing a schema file is not required.
- D. Correct.
Correct. Enabling explainable AI during training provides insights into feature importance, which can help interpret the model's predictions and improve performance.
- E. Correct.
Correct. AutoML Tables requires a clear target column (the label) and benefits from clean data (e.g., no missing values) to train accurate models.