How to build a Vertex AI pipeline for XGBoost with model comparison?

You are developing a training pipeline for a new XGBoost classification model based on tabular data. The data is stored in a BigQuery table. You need to complete the following steps: 1. Randomly split the data into training and evaluation datasets in a 65/35 ratio 2. Conduct feature engineering 3. Obtain metrics for the evaluation dataset 4. Compare models trained in different pipeline executions How should you execute these steps?

1. Using Vertex AI Pipelines, add a component to divide the data into training and evaluation sets, and add another component for feature engineering. Source Reference Answer
1. Using Vertex AI Pipelines, add a component to divide the data into training and evaluation sets, and add another component for feature engineering.
1. In BigQuery ML, use the CREATE MODEL statement with BOOSTED_TREE_CLASSIFIER as the model type and use BigQuery to handle the data splits.
1. In BigQuery ML, use the CREATE MODEL statement with BOOSTED_TREE_CLASSIFIER as the model type and use BigQuery to handle the data splits.

Community Votes

100%

100% of anonymous learners picked answer A. Votes are pick records left by other test-takers — they are not the verified answer.

Community Insight

This question tests the ability to select the right Google Cloud services for end-to-end ML workflows, specifically distinguishing between Vertex AI Experiments (for model comparison) and Vertex ML Metadata, while avoiding the trap of using BigQuery ML for complex pipeline orchestration.

To build a scalable XGBoost training pipeline with custom feature engineering and model comparison, use Vertex AI Pipelines for orchestration and Vertex AI Experiments for tracking metrics across runs. The community agrees that BigQuery ML is insufficient for complex pipeline orchestration and that Vertex AI Experiments offer superior model comparison capabilities compared to ML Metadata alone.

Choosing BigQuery ML (Options C and D) is a common mistake because, while it handles data splits and training, it lacks the robust orchestration and advanced model comparison features required for a full pipeline workflow.

Community Discussion (8 comments)

pikachu007 👍 8 Selected: A

Option B: While Vertex ML Metadata provides artifact lineage, it's less comprehensive for model comparison than Experiments. Options C and D: BigQuery ML is powerful for in-database model training, but it has limitations in pipeline orchestration, complex feature engineering, and detailed model comparison features, making it less suitable for this scenario.

wences 👍 1 Selected: A

Can anyone give a good reason for the answers without using ChatGPT or Gemini?

tardigradum 👍 1 Selected: A

BQ ML falls a bit short when it comes to building pipelines that include feature engineering and experiment comparison (it's better to use Vertex Pipelines and do the comparisons using Vertex Experiments).

fitri001 👍 1 Selected: A

Flexibility and Control: Vertex AI Pipelines allow you to define a custom pipeline with separate components for data splitting, feature engineering, and XGBoost training using your preferred libraries (like BigQueryClient and xgboost). This provides more control and customization compared to BigQuery ML's limited model types and functionality. Feature Engineering and Data Splitting: Separate components enable clear separation of concerns and potentially parallel execution for efficiency. Autologging and Model Comparison: Vertex AI autologging simplifies capturing evaluation metrics during training. Vertex AI Experiments offer a centralized interface to compare metrics across different pipeline runs (potentially with varying hyperparameter configurations).

pinimichele01 👍 1 Selected: A

see b1a8fae

omermahgoub 👍 1 Selected: A

A: Leverage Vertex AI Pipelines and Experiments

guilhermebutzke 👍 2 Selected: A

My Answer: A A: CORRECT: It involves proper data splitting into training and evaluation sets and conducting feature engineering within the pipeline, fulfilling steps 1 and 2. Enabling autologging of metrics ensures that you can track and compare the performance of different model executions, fulfilling step 3. B: Not Correct: Better use Vertex AI Experiments C and D: Not Correct: BigQuery ML lacks functionalities for comparing models across pipeline runs. You would need to rely on external tools or custom scripts to extract and compare evaluation metrics, making the process less streamlined.

b1a8fae 👍 3 Selected: A

Compare models in different pipeline executions -> go for Vertex AI experiments

Comments & Corrections

No comments yet — spotted an error or have a note? Share it below.

Expert Analysis

Why the Answer Is Correct

Option A correctly utilizes Vertex AI Pipelines to orchestrate the data splitting and feature engineering components. Crucially, it leverages Vertex AI Experiments to log metrics and visualize comparisons across multiple pipeline executions, fulfilling all four requirements of the scenario.

Why the Other Options Are Wrong

Option B suggests using Vertex ML Metadata for comparison, which provides lineage but is less effective for direct performance comparison than Vertex AI Experiments. Options C and D rely on BigQuery ML; while efficient for training, they do not support the complex, multi-step pipeline orchestration or the granular experiment tracking needed for this specific workflow.

Community Comment Notes

Commenters emphasized that the requirement to "compare models trained in different pipeline executions" is the deciding factor that points specifically to Vertex AI Experiments. Additionally, multiple users noted that BigQuery ML is limited regarding pipeline orchestration and custom feature engineering integration compared to Vertex AI Pipelines.

Official Reference

Exam Strategy

When questions ask to "compare models" across different runs, immediately look for Vertex AI Experiments as the solution. Remember that while BigQuery ML is great for training, Vertex AI Pipelines is required for complex, multi-step workflows involving custom feature engineering.

Related Analysis

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide