PMLE Architecting Low-Code ML Solutions Practice Question

A retail company wants to build a low-code ML solution to predict customer lifetime value (CLV) using historical transaction data stored in BigQuery. They have limited ML expertise and want to use BigQuery ML. Which two steps are necessary to train and evaluate a model using BigQuery ML? (Choose two.)

⚠ Common exam trap

The trap here is thinking that data must be manually split or that model export is required for training; BigQuery ML handles splitting automatically and export is only for external deployment.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Use ML.EVALUATE to assess the model's performance on a test dataset.

Training a model in BigQuery ML requires the CREATE MODEL statement, which defines and trains the model using SQL. Evaluation is performed with ML.EVALUATE to compute metrics like RMSE or accuracy. These two steps are essential for building and assessing a model. Other steps like exporting or predicting are optional or occur after evaluation, and manual data splitting is unnecessary due to automatic splitting.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Use ML.PREDICT to generate predictions on new data.
Why it's wrong here
ML.PREDICT is used to generate predictions after training and evaluation. While it is a common step in the ML lifecycle, the question specifically asks for necessary steps to train and evaluate a model. Prediction is not required for training or evaluation. Therefore, it is not one of the two necessary steps for this scenario.
✗
Manually split the data into training and test sets using a CREATE TABLE statement.
Why it's wrong here
While data splitting is important, BigQuery ML automatically splits data into training and evaluation sets when you use the CREATE MODEL statement with the DATA_SPLIT_METHOD option. Manual splitting via CREATE TABLE is not necessary and adds extra steps. It is not a required step for training and evaluating a model in BigQuery ML, and it may introduce errors if not done correctly.
✓
Use ML.EVALUATE to assess the model's performance on a test dataset.
Why this is correct
ML.EVALUATE is a function that computes evaluation metrics for a trained model, such as RMSE for regression. It is essential to assess model performance and ensure it meets business needs. You run it against a dataset not used in training. This step is necessary for model validation and is performed via SQL, maintaining the low-code approach.
✗
Export the model to a TensorFlow SavedModel for deployment.
Why it's wrong here
Exporting to TensorFlow SavedModel is not required for training and evaluating a model in BigQuery ML. It is an optional step for deploying the model outside BigQuery, such as to Vertex AI. The question asks for necessary steps to train and evaluate within BigQuery ML, so this is not needed. It also adds complexity, contradicting the low-code goal.
✓
Create a model using the CREATE MODEL statement with the appropriate model type.
Why this is correct
The CREATE MODEL statement is fundamental in BigQuery ML to define and train a model. You specify the model type, such as LINEAR_REG for regression, and provide the training data via a query. This step is necessary to initiate training. Without it, no model is created. It is a core low-code operation that uses SQL, aligning with the team's skills.

Go deeper

PMLE Architecting Low-Code ML Solutions Practice Question

Why each option matters

Option-by-option breakdown

Related to this question