easyMultiple Choice
PMLE Practice Question: A machine learning engineer is exporting a…
A machine learning engineer is exporting a trained model from Vertex AI Training to the Model Registry. Which artifact should they upload as the model artifact?
⚠ Common exam trap
Google Cloud often tests the distinction between training artifacts (checkpoints, code) and deployable model artifacts, trapping candidates who confuse a checkpoint (used for resuming training) with a final, serving-ready model.
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
✓
The saved model directory containing the model file(s) and any custom dependencies.
When exporting a trained model from Vertex AI Training to the Model Registry, the correct artifact is the saved model directory that contains the model file(s) (e.g., SavedModel format for TensorFlow, model.pkl for scikit-learn) along with any custom dependencies required for serving. This ensures the model can be deployed consistently to endpoints or batch predictions, as the Model Registry expects a self-contained artifact that includes both the model binary and its runtime dependencies.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
The saved model directory containing the model file(s) and any custom dependencies.
Why this is correct
Vertex AI Model Registry expects a SavedModel directory holding the serialised graph, variables and assets, so custom dependencies travel with it. Uploading only weights or a checkpoint omits the serving signature, preventing deployment to an Endpoint.
- ✗
Only the model checkpoint file (.ckpt or .h5).
Why it's wrong here
A bare checkpoint holds weights but omits the saved model format, signatures and preprocessing that Vertex AI serving requires to load and run inference. It is tempting because checkpoints are the training output, and would be correct for resuming training, but the registry needs an exportable SavedModel or equivalent.
- ✗
The entire training directory including training code and logs.
Why it's wrong here
The training directory bundles code, logs and checkpoints, none of which form a loadable model for Vertex AI serving. It is tempting as a complete provenance record, and would suit an experiment tracking or reproducibility archive, but the Model Registry requires a single artifact the prediction container can load directly.
- ✗
A zip file of the training source code.
Why it's wrong here
Vertex AI Model Registry expects a serialised model artifact that the serving container can load, not training source code, which cannot produce predictions. Uploading code is tempting because it documents how the model was built, and would be right for a training pipeline component or notebook, but not as the registered model artifact.
Go deeper
Related to this question
About these practice questions
This PMLE question is part of Courseiva's 775-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PMLE practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PMLE exam.