About This Domain
Domain 2 — Data Preparation — accounts for ~20% of the PMLE certification exam. This domain evaluates your understanding of data cleaning and validation, dataflow for preprocessing, bigquery for feature computation, and related concepts. Data processing for ML. To pass this section you need practical knowledge of how these services and patterns work together in real-world architectures.
What You'll Be Tested On
- Data cleaning and validation
- Dataflow for preprocessing
- BigQuery for feature computation
- Handling imbalanced datasets
Key Google Cloud Services in This Domain
Study Strategy for Domain 2
This domain represents ~20% of the total exam, making it a significant scoring area. Balance theoretical study with hands-on practice.
Exam Tips for Domain 2
Use TFX (TensorFlow Extended) for production ML pipelines.
Frequently Asked Questions
How many questions on the PMLE exam come from Domain 2?
Domain 2 (Data Preparation) makes up ~20% of the PMLE exam, approximately 16 questions.
What services should I focus on for Domain 2?
Key services include Data Preparation.
How should I prepare for Data Preparation questions?
Start by reviewing the key topics listed above, then practice with domain-specific questions. Focus on understanding real-world scenarios.
What's the best order to study the PMLE domains?
Many candidates start with the highest-weighted domains first: Architecting ML Solutions (~20%), Data Preparation (~20%), Feature Engineering (~21%), Training Models (~20%), Serving and Scaling (~19%).
Practice Domain 2 Questions
Test your knowledge of Data Preparation with practice questions from our PMLE question bank.
Start Practice Quiz →