Google Professional Machine Learning Engineer Practice Test 49
Question 480
During a controlled replay of the data and model handoff to a second engineering team, the following condition occurs: A Colab Enterprise notebook crashes when an engineer loads a multi-terabyte BigQuery table into one pandas DataFrame to compute simple counts. The handoff must preserve privacy, lineage and testable behavior. Which diagnostic check should be performed FIRST to isolate the described cause?
- Inspect Python runtime versions, pip dependency trees, package hashes and notebook execution metadata for both runs.
- Inspect the labeling payload schema, data classification, annotator roles and fields needed to complete the labeling task.
- Restart the runtime, run the notebook top to bottom and compare variable initialization with committed source revision.
- Inspect query result sizes, runtime memory, pandas allocations and which aggregations can run in BigQuery SQL.
Explanation
Correct Answer
D:Inspect query result sizes, runtime memory, pandas allocations and which aggregations can run in BigQuery SQL. This inspects evidence relevant to notebook memory exhaustion on full data extraction. Single-machine notebook memory is limited; warehouse operations should execute near large warehouse-resident data. After identifying the cause, the matching correction is: Push aggregations into BigQuery or stream bounded samples to the notebook rather than materializing the whole table.
Incorrect Answers
A:Inspect Python runtime versions, pip dependency trees, package hashes and notebook execution metadata for both runs. This examines notebook dependency lock reproducibility rather than the specific mechanism reported here. It is not the decisive first check for notebook memory exhaustion on full data extraction. The relevant evidence is: Inspect query result sizes, runtime memory, pandas allocations and which aggregations can run in BigQuery SQL.
B:Inspect the labeling payload schema, data classification, annotator roles and fields needed to complete the labeling task. This examines data minimization for labeling inputs rather than the specific mechanism reported here. It is not the decisive first check for notebook memory exhaustion on full data extraction. The relevant evidence is: Inspect query result sizes, runtime memory, pandas allocations and which aggregations can run in BigQuery SQL.
C:Restart the runtime, run the notebook top to bottom and compare variable initialization with committed source revision. This examines notebook transient state and execution order rather than the specific mechanism reported here. It is not the decisive first check for notebook memory exhaustion on full data extraction. The relevant evidence is: Inspect query result sizes, runtime memory, pandas allocations and which aggregations can run in BigQuery SQL.
Question 481
During a staged restoration test at a secondary serving deployment, the following condition occurs: A team deploys a model using a mutable storage path named production-latest and cannot reconstruct which evaluation approved its weights. The team must prove an actual user-visible improvement without weakening access controls. Select TWO: the decisive diagnostic step and the targeted repair for this specific issue.
- Choose a managed batch inference or scheduled container batch workflow instead of an always-on online endpoint.
- Promote an explicitly approved model version from Agent Platform Model Registry and record deployment lineage.
- Review payload size distributions, request duration, response timeout policies and tolerance for deferred completion.
- Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata.
- Configure an appropriate request-concurrency or workload-driven scaling policy with resource and cooldown limits.
Explanation
Correct Answers
B:Promote an explicitly approved model version from Agent Platform Model Registry and record deployment lineage. This is the targeted corrective response for model registry version provenance. Mutable nicknames do not establish immutable model provenance for controlled production releases. Verify the diagnosis using: Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata. Confirm success using: The running endpoint resolves to a traceable evaluated package and cannot silently switch weights when an alias moves.
D:Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata. This inspects evidence relevant to model registry version provenance. Mutable nicknames do not establish immutable model provenance for controlled production releases. After identifying the cause, the matching correction is: Promote an explicitly approved model version from Agent Platform Model Registry and record deployment lineage.
Incorrect Answers
A:Choose a managed batch inference or scheduled container batch workflow instead of an always-on online endpoint. This is an action or diagnostic relevant to batch versus online inference schedule, not one of the required two steps. In this scenario verify: Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata. Then correct: Promote an explicitly approved model version from Agent Platform Model Registry and record deployment lineage.
C:Review payload size distributions, request duration, response timeout policies and tolerance for deferred completion. This is an action or diagnostic relevant to online prediction timeout and batch fallback, not one of the required two steps. In this scenario verify: Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata. Then correct: Promote an explicitly approved model version from Agent Platform Model Registry and record deployment lineage.
E:Configure an appropriate request-concurrency or workload-driven scaling policy with resource and cooldown limits. This is an action or diagnostic relevant to throughput-based autoscaling versus cpu metric, not one of the required two steps. In this scenario verify: Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata. Then correct: Promote an explicitly approved model version from Agent Platform Model Registry and record deployment lineage.
Question 482
While reproducing a training orchestration incident in isolated staging, the following condition occurs: A model registry entry lacks the upstream version of the training feature table, preventing investigators from reconstructing a reported prediction. The team requires an idempotent pipeline behavior and traceable evidence. Which underlying engineering principle explains why the original approach fails?
- Model retraining reproducibility requires stable inputs rather than unversioned reads of an evolving table.
- Model weights alone do not capture the data lineage necessary to reproduce and investigate predictions.
- Two independent preprocessing implementations can diverge even when their initial training metrics look correct.
- A schema existing does not imply data quality; pipeline gates must enforce business completeness rules.
Explanation
Correct Answer
B:Model weights alone do not capture the data lineage necessary to reproduce and investigate predictions. This explains the mechanism behind model feature lineage for audit and why the described failure is possible. A discriminating investigation is: Inspect ML Metadata execution links, table snapshots, transformations and model version lineage. The corresponding repair is: Record governed dataset snapshot and transformation versions as pipeline inputs linked to the resulting model artifact.
Incorrect Answers
A:Model retraining reproducibility requires stable inputs rather than unversioned reads of an evolving table. This principle describes data version reproducibility in ct jobs, which is not the cause of the observed model feature lineage for audit failure. The controlling explanation is: Model weights alone do not capture the data lineage necessary to reproduce and investigate predictions.
C:Two independent preprocessing implementations can diverge even when their initial training metrics look correct. This principle describes training serving preprocessing artifact sharing, which is not the cause of the observed model feature lineage for audit failure. The controlling explanation is: Model weights alone do not capture the data lineage necessary to reproduce and investigate predictions.
D:A schema existing does not imply data quality; pipeline gates must enforce business completeness rules. This principle describes data validation threshold gate, which is not the cause of the observed model feature lineage for audit failure. The controlling explanation is: Model weights alone do not capture the data lineage necessary to reproduce and investigate predictions.
Question 483
During a staged restoration test at a secondary serving deployment, the following condition occurs: A deployment grants every project member broad model prediction permissions merely to make a limited partner integration work. The team must prove an actual user-visible improvement without weakening access controls. Which result is the strongest postchange evidence of success?
- Large documents finish without client timeout and consumers retrieve signed results using recorded job IDs.
- Regional clients experience lower p95 end-to-end latency without changes to model compute quality.
- The approved integration continues to work while unrelated accounts receive permission denied on inference calls.
- Latency and memory decline on target hardware while validation degradation stays within the approved threshold.
Explanation
Correct Answer
C:The approved integration continues to work while unrelated accounts receive permission denied on inference calls. This is specific acceptance evidence for public endpoint iam oversharing. Endpoint invoker permissions should follow least privilege, rather than expanding access for convenience. It confirms the controlled correction: Restrict invoke access to the necessary principal and endpoint, using appropriate short-lived authentication. This matters here because during a staged restoration test at a secondary serving deployment, the following condition occurs: A deployment….
Incorrect Answers
A:Large documents finish without client timeout and consumers retrieve signed results using recorded job IDs. This result could demonstrate success for online prediction timeout and batch fallback but not for the stated public endpoint iam oversharing issue. The direct test is: The approved integration continues to work while unrelated accounts receive permission denied on inference calls.
B:Regional clients experience lower p95 end-to-end latency without changes to model compute quality. This result could demonstrate success for serving region proximity versus model compute but not for the stated public endpoint iam oversharing issue. The direct test is: The approved integration continues to work while unrelated accounts receive permission denied on inference calls.
D:Latency and memory decline on target hardware while validation degradation stays within the approved threshold. This result could demonstrate success for inference optimization using quantization but not for the stated public endpoint iam oversharing issue. The direct test is: The approved integration continues to work while unrelated accounts receive permission denied on inference calls.
Question 484
During a regulated MLOps pipeline deployment, the following condition occurs: A build pipeline deploys an arbitrary model tag without tying it to the source revision and the approved evaluation report. The workflow must be repeatable and protect the last approved production model. What should the engineer inspect before selecting a correction?
- Inspect Cloud Build substitutions, commit SHAs, model registry versions, signed image digests and approval records.
- Inspect request schema, container route behavior, test fixtures and CI stage dependency graph.
- Compare serialized preprocessing artifacts, schema versions, train-time transformations and live request feature IDs.
- Review champion model retention, candidate metrics, change approvals and automatic rollback triggers.
Explanation
Correct Answer
A:Inspect Cloud Build substitutions, commit SHAs, model registry versions, signed image digests and approval records. This inspects evidence relevant to cloud build release version promotion. Unpinned tags do not establish a trustworthy relationship between code, model artifacts and production rollout. After identifying the cause, the matching correction is: Promote explicit immutable model and image versions with versioned release manifests and review gates.
Incorrect Answers
B:Inspect request schema, container route behavior, test fixtures and CI stage dependency graph. This examines continuous delivery integration contract tests rather than the specific mechanism reported here. It is not the decisive first check for cloud build release version promotion. The relevant evidence is: Inspect Cloud Build substitutions, commit SHAs, model registry versions, signed image digests and approval records.
C:Compare serialized preprocessing artifacts, schema versions, train-time transformations and live request feature IDs. This examines training serving preprocessing artifact sharing rather than the specific mechanism reported here. It is not the decisive first check for cloud build release version promotion. The relevant evidence is: Inspect Cloud Build substitutions, commit SHAs, model registry versions, signed image digests and approval records.
D:Review champion model retention, candidate metrics, change approvals and automatic rollback triggers. This examines continuous training rollback protection rather than the specific mechanism reported here. It is not the decisive first check for cloud build release version promotion. The relevant evidence is: Inspect Cloud Build substitutions, commit SHAs, model registry versions, signed image digests and approval records.
Question 485
During a regulated MLOps pipeline deployment, the following condition occurs: An endpoint calculates categorical encodings in application code while the training pipeline uses a separate mapping, creating drift after a feature update. The workflow must be repeatable and protect the last approved production model. What implementation change resolves the specific issue without substituting an unrelated fix?
- Add a data-validation component that quarantines nonconforming inputs and fails downstream training until reviewed.
- Add explicit upstream validation and evaluation dependencies before registration and deployment components.
- Package the fitted preprocessing component with the model and apply the same validated transform in both paths.
- Create immutable versioned input snapshots or partition manifests and record them in training metadata.
Explanation
Correct Answer
C:Package the fitted preprocessing component with the model and apply the same validated transform in both paths. This is the targeted corrective response for training serving preprocessing artifact sharing. Two independent preprocessing implementations can diverge even when their initial training metrics look correct. Verify the diagnosis using: Compare serialized preprocessing artifacts, schema versions, train-time transformations and live request feature IDs. Confirm success using: Representative raw records produce equal feature vectors offline and online, including unseen-category behavior. This matters here because an endpoint calculates categorical encodings in application code while the training pipeline uses a separate….
Incorrect Answers
A:Add a data-validation component that quarantines nonconforming inputs and fails downstream training until reviewed. This action addresses data validation threshold gate, not the failure in the described training serving preprocessing artifact sharing case. The targeted correction is: Package the fitted preprocessing component with the model and apply the same validated transform in both paths.
B:Add explicit upstream validation and evaluation dependencies before registration and deployment components. This action addresses pipeline component dependency order, not the failure in the described training serving preprocessing artifact sharing case. The targeted correction is: Package the fitted preprocessing component with the model and apply the same validated transform in both paths.
D:Create immutable versioned input snapshots or partition manifests and record them in training metadata. This action addresses data version reproducibility in ct jobs, not the failure in the described training serving preprocessing artifact sharing case. The targeted correction is: Package the fitted preprocessing component with the model and apply the same validated transform in both paths.
Question 486
Within a jointly maintained Google Cloud data science workspace, the following condition occurs: Two experiments use the same model configuration but show different accuracy due to untracked data shuffling and nondeterministic seeds. Collaborators must avoid expanding data access or breaking experiment reproducibility. Which targeted engineering change most directly addresses the failure?
- Log reproducibility settings and rerun comparable trials with fixed split and controlled randomness before attributing changes.
- Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates.
- Record input dataset versions, training runs, code revisions and output artifact links in the managed metadata and registry workflow.
- Implement the batch aggregate in scheduled BigQuery SQL when the required transformation is expressible and fits its SLA.
Explanation
Correct Answer
A:Log reproducibility settings and rerun comparable trials with fixed split and controlled randomness before attributing changes. This is the targeted corrective response for experiment random seed and data order variance. A single stochastic training run is not sufficient to distinguish algorithm improvement from sampling variation. Verify the diagnosis using: Inspect random seeds, sample ordering, framework determinism settings and environment versions recorded per run. Confirm success using: Repeated controlled runs show bounded expected variance and any real improvement exceeds run-to-run noise.
Incorrect Answers
B:Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates. This action addresses experiment metric comparison with mismatched splits, not the failure in the described experiment random seed and data order variance case. The targeted correction is: Log reproducibility settings and rerun comparable trials with fixed split and controlled randomness before attributing changes.
C:Record input dataset versions, training runs, code revisions and output artifact links in the managed metadata and registry workflow. This action addresses ml metadata lineage for production model, not the failure in the described experiment random seed and data order variance case. The targeted correction is: Log reproducibility settings and rerun comparable trials with fixed split and controlled randomness before attributing changes.
D:Implement the batch aggregate in scheduled BigQuery SQL when the required transformation is expressible and fits its SLA. This action addresses bigquery versus dataflow preparation selection, not the failure in the described experiment random seed and data order variance case. The targeted correction is: Log reproducibility settings and rerun comparable trials with fixed split and controlled randomness before attributing changes.
Question 487
During a production ML inference release on Google Cloud, the following condition occurs: A GPU inference server shows low CPU use but increasing queue length and slow responses under request bursts; the only scaling rule watches CPU. The service must meet its agreed quality and latency requirements. Which statement identifies the primary reason for this failure?
- Coercing identifier types can change keys, particularly when numeric-looking strings have meaningful leading zeros.
- Identity permission is not proof of network reachability; private serving also requires correct routing and name resolution.
- A valid container image is not necessarily compatible with an inference service's HTTP startup and health contract.
- CPU usage can be weakly related to accelerator or queue-bound inference workloads.
Explanation
Correct Answer
D:CPU usage can be weakly related to accelerator or queue-bound inference workloads. This explains the mechanism behind throughput-based autoscaling versus cpu metric and why the described failure is possible. A discriminating investigation is: Compare request rate, queue depth, GPU utilization, per-instance concurrency and p95 response latency. The corresponding repair is: Configure an appropriate request-concurrency or workload-driven scaling policy with resource and cooldown limits.
Incorrect Answers
A:Coercing identifier types can change keys, particularly when numeric-looking strings have meaningful leading zeros. This principle describes feature online entity key mismatch, which is not the cause of the observed throughput-based autoscaling versus cpu metric failure. The controlling explanation is: CPU usage can be weakly related to accelerator or queue-bound inference workloads.
B:Identity permission is not proof of network reachability; private serving also requires correct routing and name resolution. This principle describes private vpc routing failure for serving, which is not the cause of the observed throughput-based autoscaling versus cpu metric failure. The controlling explanation is: CPU usage can be weakly related to accelerator or queue-bound inference workloads.
C:A valid container image is not necessarily compatible with an inference service's HTTP startup and health contract. This principle describes inference container health interface mismatch, which is not the cause of the observed throughput-based autoscaling versus cpu metric failure. The controlling explanation is: CPU usage can be weakly related to accelerator or queue-bound inference workloads.
Question 488
At a Google Cloud analytics team preparing a governed pilot, the following condition occurs: A demand team needs a 21-day rolling forecast but creates a classification model on randomly shuffled dates and never defines the forecast horizon. The team must satisfy a reproducible low-code AI assessment. Which outcome most directly demonstrates a successful, measurable fix?
- A controlled raw record yields the same transformed feature representation for evaluation and production SQL inference.
- The chosen version meets task quality, inference cost and license rules on a documented representative evaluation set.
- Twenty-one future daily estimates are produced with reported prediction intervals and error is scored on forward dates.
- An unseen labeled cohort receives probabilities for the intended positive class with a calibrated operating threshold.
Explanation
Correct Answer
C:Twenty-one future daily estimates are produced with reported prediction intervals and error is scored on forward dates. This is specific acceptance evidence for bigquery arima_plus forecast horizon. Forecasting requires time-ordered histories and explicitly defined future steps, not shuffled categorical labels. It confirms the controlled correction: Use a suitable ARIMA_PLUS time-series model with timestamp and data columns and evaluate ML.FORECAST on rolling historical cutoffs.
Incorrect Answers
A:A controlled raw record yields the same transformed feature representation for evaluation and production SQL inference. This result could demonstrate success for bigquery ml transform serving parity but not for the stated bigquery arima_plus forecast horizon issue. The direct test is: Twenty-one future daily estimates are produced with reported prediction intervals and error is scored on forward dates.
B:The chosen version meets task quality, inference cost and license rules on a documented representative evaluation set. This result could demonstrate success for model garden task and license selection but not for the stated bigquery arima_plus forecast horizon issue. The direct test is: Twenty-one future daily estimates are produced with reported prediction intervals and error is scored on forward dates.
D:An unseen labeled cohort receives probabilities for the intended positive class with a calibrated operating threshold. This result could demonstrate success for bigquery ml logistic model for conversion probability but not for the stated bigquery arima_plus forecast horizon issue. The direct test is: Twenty-one future daily estimates are produced with reported prediction intervals and error is scored on forward dates.
Question 489
While preparing a production training workflow on Gemini Enterprise Agent Platform, the following condition occurs: A team wants to tune an image-generation foundation model using scraped brand imagery without verifying usage rights or consent. Engineers need an auditable, resource-aware correction to the issue. Select TWO: the decisive diagnostic step and the targeted repair for this specific issue.
- Use licensed and permitted training images with appropriate governance reviews before any supported tuning job.
- Reduce the learning rate and use justified parameter-efficient tuning or partial freezing with a controlled evaluation set.
- Inspect gradient ranges, loss-scaling logs, overflow events and the affected operations with test batches.
- Check dataset licenses, provenance, consent policies, safety requirements and the model's supported tuning route.
- Use supported structured output and schema validation with graceful retry or rejection for nonconforming responses.
Explanation
Correct Answers
A:Use licensed and permitted training images with appropriate governance reviews before any supported tuning job. This is the targeted corrective response for fine-tuning image generator data rights. Fine-tuning technical feasibility does not confer rights to reuse restricted training images or ignore model safety. Verify the diagnosis using: Check dataset licenses, provenance, consent policies, safety requirements and the model's supported tuning route. Confirm success using: The training manifest contains traceable permitted sources and safety tests detect known prohibited outputs. This matters here because while preparing a production training workflow on Gemini Enterprise Agent Platform, the following condition occurs:….
D:Check dataset licenses, provenance, consent policies, safety requirements and the model's supported tuning route. This inspects evidence relevant to fine-tuning image generator data rights. Fine-tuning technical feasibility does not confer rights to reuse restricted training images or ignore model safety. After identifying the cause, the matching correction is: Use licensed and permitted training images with appropriate governance reviews before any supported tuning job. This matters here because while preparing a production training workflow on Gemini Enterprise Agent Platform, the following condition occurs:….
Incorrect Answers
B:Reduce the learning rate and use justified parameter-efficient tuning or partial freezing with a controlled evaluation set. This is an action or diagnostic relevant to fine-tuning catastrophic forgetting, not one of the required two steps. In this scenario verify: Check dataset licenses, provenance, consent policies, safety requirements and the model's supported tuning route. Then correct: Use licensed and permitted training images with appropriate governance reviews before any supported tuning job.
C:Inspect gradient ranges, loss-scaling logs, overflow events and the affected operations with test batches. This is an action or diagnostic relevant to mixed precision overflow diagnosis, not one of the required two steps. In this scenario verify: Check dataset licenses, provenance, consent policies, safety requirements and the model's supported tuning route. Then correct: Use licensed and permitted training images with appropriate governance reviews before any supported tuning job.
E:Use supported structured output and schema validation with graceful retry or rejection for nonconforming responses. This is an action or diagnostic relevant to constrained generative json outputs, not one of the required two steps. In this scenario verify: Check dataset licenses, provenance, consent policies, safety requirements and the model's supported tuning route. Then correct: Use licensed and permitted training images with appropriate governance reviews before any supported tuning job.
Question 490
During a postdeployment AI safety and reliability review, the following condition occurs: An LLM summarizer produces fluent answers containing unsupported policy details, but the team measures only response latency and vocabulary similarity. The model owner must provide repeatable evidence rather than a cosmetic dashboard change. Which diagnostic check should be performed FIRST to isolate the described cause?
- Inspect attachment ingestion, trust boundaries, malicious prompt examples and deployed Model Armor filter results.
- Examine alert history, day-of-week distributions, false-positive cost and baseline segmentation.
- Inspect retrieved evidence, claim-level factual consistency, reference answers and human review samples.
- Compare live and offline feature SQL, window definitions, sample entity values and prediction-time cutoffs.
Explanation
Correct Answer
C:Inspect retrieved evidence, claim-level factual consistency, reference answers and human review samples. This inspects evidence relevant to generative ai factual grounding evaluation. Fluent text or semantic similarity is not proof an answer is supported by the retrieved source material. After identifying the cause, the matching correction is: Add groundedness and factual correctness checks to a representative gen AI evaluation suite and tune retrieval or prompting accordingly. This matters here because an LLM summarizer produces fluent answers containing unsupported policy details, but the team measures only response….
Incorrect Answers
A:Inspect attachment ingestion, trust boundaries, malicious prompt examples and deployed Model Armor filter results. This examines model armor untrusted content screening rather than the specific mechanism reported here. It is not the decisive first check for generative ai factual grounding evaluation. The relevant evidence is: Inspect retrieved evidence, claim-level factual consistency, reference answers and human review samples.
B:Examine alert history, day-of-week distributions, false-positive cost and baseline segmentation. This examines model monitoring false-alert seasonality rather than the specific mechanism reported here. It is not the decisive first check for generative ai factual grounding evaluation. The relevant evidence is: Inspect retrieved evidence, claim-level factual consistency, reference answers and human review samples.
D:Compare live and offline feature SQL, window definitions, sample entity values and prediction-time cutoffs. This examines training-serving feature skew detection rather than the specific mechanism reported here. It is not the decisive first check for generative ai factual grounding evaluation. The relevant evidence is: Inspect retrieved evidence, claim-level factual consistency, reference answers and human review samples. This matters here because an LLM summarizer produces fluent answers containing unsupported policy details, but the team measures only response….