Databricks Certified Machine Learning Professional Practice Test 100
Question 990
Consider the following fictional Databricks production case: A team switches a SparkML training stage to Ray without verifying that its chosen estimator supports Ray's distributed fit contracts. What consequence most specifically follows from leaving the issue unchanged?
- A single skewed group repeatedly kills workers despite healthy median group runtime.
- The rewritten job runs only on one worker or fails during model serialization.
- Every replica hits device OOM even though aggregate cluster memory appears ample.
- The apparent improvement is a data-split artifact rather than a better model.
Explanation
Correct Answer
B:This consequence follows from the specific unresolved condition. The rewritten job runs only on one worker or fails during model serialization. The causal constraint is: Selecting Ray over Spark must be based on the actual training library and distributed API, not the existence of a cluster.
Incorrect Answers
A:That direct consequence is directed at a different scaling and tuning situation: One enterprise account holds 80% of historical events, and grouped pandas training for that account exceeds executor memory while other accounts finish. The more immediate risk here is the rewritten job runs only on one worker or fails during model serialization.
C:That direct consequence is directed at a different scaling and tuning situation: An image model needs 90 GB accelerator memory for one training replica while each available GPU has 24 GB and engineers propose only adding more identical nodes. The more immediate risk here is the rewritten job runs only on one worker or fails during model serialization.
D:That direct consequence is directed at a different scaling and tuning situation: A hyperparameter team compares two Optuna studies with different train/validation splits and chooses the lower loss as though both metrics were directly comparable. The more immediate risk here is the rewritten job runs only on one worker or fails during model serialization.
Question 991
The engineering team encounters the following issue: A regulated company changes deployed Python notebook code directly in production to correct a model transform while its test environment runs a different untracked revision. What direct consequence should the owner prioritize if no fix is made?
- The wrong experiment artifact may be used in production because lifecycle stages are disconnected.
- The live model logic diverges from tested code and cannot be reproduced during an incident.
- The artifact can disappear with an account change or be overwritten without reliable provenance.
- A user can access or mutate a deployed model despite restricted tracking visibility.
Explanation
Correct Answer
B:This consequence follows from the specific unresolved condition. The live model logic diverges from tested code and cannot be reproduced during an incident. The causal constraint is: A deploy-code strategy promotes a tested, versioned code artifact through environments rather than editing production independently.
Incorrect Answers
A:That direct consequence is directed at a different model lifecycle management situation: A team records experiments but has no controlled path from a selected registered model to scheduled scoring jobs. The more immediate risk here is the live model logic diverges from tested code and cannot be reproduced during an incident.
C:That direct consequence is directed at a different model lifecycle management situation: An engineer stores model weights in a personal workspace folder and treats that path as a production artifact registry with audit history. The more immediate risk here is the live model logic diverges from tested code and cannot be reproduced during an incident.
D:That direct consequence is directed at a different model lifecycle management situation: A platform team proposes using MLflow Tracking alone to enforce production endpoint authorization, ignoring Unity Catalog and service permissions. The more immediate risk here is the live model logic diverges from tested code and cannot be reproduced during an incident.
Question 992
Review this production model incident: A new merchant group contributes only 12 observations, yet a monitor reports a large categorical divergence from a baseline of 50,000 and the team assumes the result is definitive. What evidence would best distinguish the actual fault from neighboring ML concerns?
- Compare segment filters, join population and sampling criteria for both sides of the statistic.
- Read both window sizes and the supported categorical comparison statistic for the small slice.
- Inspect effective owner permissions, catalog/schema access and creation error logs for the target table.
- Compare feature drift trends with label-matured error rates and calibration by model version.
Explanation
Correct Answer
B:This is the most discriminating check for the observed symptoms. Read both window sizes and the supported categorical comparison statistic for the small slice. The underlying technical issue is that drift metrics on tiny subgroups require minimum-sample and uncertainty checks before decisive interpretation.
Incorrect Answers
A:That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A customer-risk monitoring job compares an unfiltered current high-risk segment to an all-customer historical baseline and interprets the difference as temporal drift. For the stated incident, the decisive observation is read both window sizes and the supported categorical comparison statistic for the small slice.
C:That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A monitor is pointed at a Unity Catalog table but the automation identity lacks the privileges needed to generate profile outputs. For the stated incident, the decisive observation is read both window sizes and the supported categorical comparison statistic for the small slice.
D:That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A risk service measures feature drift but neglects observed-label performance and continues serving an accurate-looking yet ineffective model. For the stated incident, the decisive observation is read both window sizes and the supported categorical comparison statistic for the small slice.
Question 993
The engineering team encounters the following issue: A retailer's features arrive two hours late, and offline training uses event time alone even though the online system could not have seen those late arrivals at prediction time. What corrective action should the engineer prefer?
- Add a supported duplicate-safe event processing strategy and preserve source identifiers.
- Use deterministic merge or overwrite semantics scoped to logical feature keys and verify retry behavior.
- Apply the approved availability cutoff and exclude features not accessible when the decision was made.
- Configure the supported synchronization process and freshness alerts for critical features.
Explanation
Correct Answer
C:This corrects the relevant failure boundary instead of merely hiding the symptom. Apply the approved availability cutoff and exclude features not accessible when the decision was made. The issue matters because a backtest gains information that production inference could not have used.
Incorrect Answers
A:That corrective action is directed at a different advanced feature store concepts situation: A real-time feature job updates user purchase totals without checking whether upstream transaction events are delivered twice. The relevant corrective action here is apply the approved availability cutoff and exclude features not accessible when the decision was made.
B:That corrective action is directed at a different advanced feature store concepts situation: A feature refresh writes duplicate records for each entity and logical timestamp after retries, breaking unique lookups. The relevant corrective action here is apply the approved availability cutoff and exclude features not accessible when the decision was made.
D:That corrective action is directed at a different advanced feature store concepts situation: A team publishes an online feature table but neglects its refresh schedule while the predictive model assumes values no older than 15 minutes. The relevant corrective action here is apply the approved availability cutoff and exclude features not accessible when the decision was made.
Question 994
Review this production model incident: An engineering group duplicates a validation helper across six notebooks, and two copies enforce outdated threshold values. What end-to-end verification is most persuasive here?
- All six consumers use one pinned helper version and identical threshold fixtures.
- The unit suite confirms known instants normalize consistently across supported offset formats.
- The exact saved fixture is successfully scored via the deployed endpoint with expected fields.
- The production image digest matches the tested release and reproduces expected fixture outputs.
Explanation
Correct Answer
A:This verification measures the required result rather than just process liveness. All six consumers use one pinned helper version and identical threshold fixtures. It confirms whether the targeted action succeeded: centralize versioned validation functions and test behavior under all consuming environments.
Incorrect Answers
B:That acceptance test is directed at a different validation testing situation: A date feature function treats all incoming timestamp strings as UTC even though some include explicit offsets and a unit test uses only Z-suffixed inputs. Here the actual success criterion is all six consumers use one pinned helper version and identical threshold fixtures.
C:That acceptance test is directed at a different validation testing situation: A churn feature job succeeds and a classifier achieves acceptable F1, but the deployed endpoint returns HTTP 400 because request column names differ. Here the actual success criterion is all six consumers use one pinned helper version and identical threshold fixtures.
D:That acceptance test is directed at a different validation testing situation: A release pipeline runs development unit tests against a different dependency image from the one deployed to production. Here the actual success criterion is all six consumers use one pinned helper version and identical threshold fixtures.
Question 995
A production ML review contains this situation: A risk service measures feature drift but neglects observed-label performance and continues serving an accurate-looking yet ineffective model. Which diagnostic check provides the strongest direct evidence?
- Review rolling-window statistics alongside a fixed approved baseline for the same cohort.
- Inspect prediction, model-id, event-time and label fields in the captured inference table.
- Test timestamp conversion and timezone interpretation for valid and malformed synthetic source rows.
- Compare feature drift trends with label-matured error rates and calibration by model version.
Explanation
Correct Answer
D:This is the most discriminating check for the observed symptoms. Compare feature drift trends with label-matured error rates and calibration by model version. The underlying technical issue is that input distribution and predictive outcome quality are complementary monitoring signals with different evidence needs.
Incorrect Answers
A:That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A streaming payments monitor always compares today with the previous hour, and a slow week-long distribution drift escapes because adjacent hours differ only slightly. For the stated incident, the decisive observation is compare feature drift trends with label-matured error rates and calibration by model version.
B:That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A churn model logs requests, predictions, model versions and later ground-truth labels, but monitoring is configured as a generic feature snapshot. For the stated incident, the decisive observation is compare feature drift trends with label-matured error rates and calibration by model version.
C:That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A time-series monitor uses a string-formatted date that cannot be reliably converted to timestamps and silently groups most rows into an unknown bucket. For the stated incident, the decisive observation is compare feature drift trends with label-matured error rates and calibration by model version.
Question 996
Consider the following fictional Databricks production case: A low-latency feature table uses external customer names as lookup keys; names change and are not unique. Which implementation is most appropriate for the stated incident?
- Centralize eligible feature definitions and reuse a controlled feature computation workflow where supported.
- Promote a versioned feature definition with backward-compatibility tests and controlled rollout.
- Add a supported duplicate-safe event processing strategy and preserve source identifiers.
- Use a stable governed primary key and update feature publication and serving lookup mappings.
Explanation
Correct Answer
D:This corrects the relevant failure boundary instead of merely hiding the symptom. Use a stable governed primary key and update feature publication and serving lookup mappings. The issue matters because scores pull the wrong customer's features or fail after name changes.
Incorrect Answers
A:That corrective action is directed at a different advanced feature store concepts situation: An ML pipeline computes training features in one notebook and duplicates slightly different SQL in the online pipeline for the same entities. The relevant corrective action here is use a stable governed primary key and update feature publication and serving lookup mappings.
B:That corrective action is directed at a different advanced feature store concepts situation: A team updates a shared on-demand feature function without version control and unexpectedly changes both batch retraining and online serving behavior. The relevant corrective action here is use a stable governed primary key and update feature publication and serving lookup mappings.
C:That corrective action is directed at a different advanced feature store concepts situation: A real-time feature job updates user purchase totals without checking whether upstream transaction events are delivered twice. The relevant corrective action here is use a stable governed primary key and update feature publication and serving lookup mappings.
Question 997
The engineering team encounters the following issue: A team edits a model endpoint through the UI while its CI pipeline concurrently applies an older configuration and reverts the intended traffic split. What corrective action should the engineer prefer?
- Route changes through a single approved configuration source and verify the active endpoint after each change.
- Log the supported custom PyFunc model and register it to an authorized governed model name.
- Use a properly scoped authorized credential and supported authenticated inference request.
- Remove embedded secrets and configure an approved scoped secret retrieval mechanism for the serving process.
Explanation
Correct Answer
A:This corrects the relevant failure boundary instead of merely hiding the symptom. Route changes through a single approved configuration source and verify the active endpoint after each change. The issue matters because manual and automated updates race, creating unexpected model routing during production traffic.
Incorrect Answers
B:That corrective action is directed at a different custom model serving situation: A research group wants to register a custom classification model in Unity Catalog but uses a local path as its complete model identity and has no governed model name. The relevant corrective action here is route changes through a single approved configuration source and verify the active endpoint after each change.
C:That corrective action is directed at a different custom model serving situation: A client invokes a registered custom ML model using an unauthenticated HTTP POST to the serving endpoint and assumes network access implies authorization. The relevant corrective action here is route changes through a single approved configuration source and verify the active endpoint after each change.
D:That corrective action is directed at a different custom model serving situation: An engineer logs a customer-specific API token inside the pickled PythonModel object before registering it to Unity Catalog. The relevant corrective action here is route changes through a single approved configuration source and verify the active endpoint after each change.
Question 998
The engineering team encounters the following issue: An ML project stores test files directly in each production notebook and execution of a deployed notebook inadvertently reruns expensive tests. Which technical principle most directly explains the root of the issue?
- Unit tests should verify reproducible content and stable mappings rather than shape alone.
- Test organization should separate production orchestration from the test runner and control execution stages.
- Release-stage testing must validate the actual artifact and dependency environment being promoted.
- A full ML integration test should traverse raw feature generation through training, registration, deployment and inference contracts.
Explanation
Correct Answer
B:The technical constraint is specific to the reported failure. Test organization should separate production orchestration from the test runner and control execution stages. To test it, inspect notebook entrypoints, test invocation conditions and deployment job definitions.
Incorrect Answers
A:That underlying principle is directed at a different validation testing situation: A notebook's label encoder produces a different category order depending on an unordered Python set, but tests assert only output length. The current incident instead turns on test organization should separate production orchestration from the test runner and control execution stages.
C:That underlying principle is directed at a different validation testing situation: A release pipeline runs development unit tests against a different dependency image from the one deployed to production. The current incident instead turns on test organization should separate production orchestration from the test runner and control execution stages.
D:That underlying principle is directed at a different validation testing situation: A churn feature job succeeds and a classifier achieves acceptable F1, but the deployed endpoint returns HTTP 400 because request column names differ. The current incident instead turns on test organization should separate production orchestration from the test runner and control execution stages.
Question 999
In an enterprise ML workflow, the reported issue is: A monitoring dashboard reports a statistically significant tiny difference after one million observations, and the team immediately treats it as operationally material drift. Which change addresses the real cause without undermining other requirements?
- Normalize and validate a supported timestamp field before applying the configured temporal profile.
- Define joint effect-size and operational-importance thresholds rather than escalating every low p-value.
- Add an appropriately sustained alert condition or window aggregate instead of a one-minute peak trigger.
- Separate versioned baseline storage from current data and block unintended updates.
Explanation
Correct Answer
B:This corrects the relevant failure boundary instead of merely hiding the symptom. Define joint effect-size and operational-importance thresholds rather than escalating every low p-value. The issue matters because alert fatigue grows from trivial shifts that do not affect decisions or predictions.
Incorrect Answers
A:That corrective action is directed at a different drift detection and lakehouse monitoring situation: A time-series monitor uses a string-formatted date that cannot be reliably converted to timestamps and silently groups most rows into an unknown bucket. The relevant corrective action here is define joint effect-size and operational-importance thresholds rather than escalating every low p-value.
C:That corrective action is directed at a different drift detection and lakehouse monitoring situation: A monitor alerts whenever a ratio exceeds 0.3 for a single minute, although normal traffic produces brief spikes and operations require ten-minute persistence. The relevant corrective action here is define joint effect-size and operational-importance thresholds rather than escalating every low p-value.
D:That corrective action is directed at a different drift detection and lakehouse monitoring situation: A baseline table is overwritten every night with the current day's data before a drift calculation, making the comparison always look unchanged. The relevant corrective action here is define joint effect-size and operational-importance thresholds rather than escalating every low p-value.
Question 1000
Review this production model incident: A customer segmentation project promotes a scoring job but omits a database migration required by its new feature schema. Which mitigation resolves the defect rather than concealing it?
- Record end-to-end lineage and immutable versions at each necessary lifecycle boundary.
- Add a backward-compatible schema migration and gated deployment order to the release workflow.
- Apply the correct governance controls separately to experiments, registered models and serving endpoints.
- Bind training outputs to registered immutable versions and promote them through governed jobs or serving endpoints.
Explanation
Correct Answer
B:This corrects the relevant failure boundary instead of merely hiding the symptom. Add a backward-compatible schema migration and gated deployment order to the release workflow. The issue matters because a healthy deployed job crashes on missing fields or silently reads wrong defaults.
Incorrect Answers
A:That corrective action is directed at a different model lifecycle management situation: An organization has reliable model serving but never archives feature definitions or experiment parameters and cannot reproduce predictions after drift. The relevant corrective action here is add a backward-compatible schema migration and gated deployment order to the release workflow.
C:That corrective action is directed at a different model lifecycle management situation: A platform team proposes using MLflow Tracking alone to enforce production endpoint authorization, ignoring Unity Catalog and service permissions. The relevant corrective action here is add a backward-compatible schema migration and gated deployment order to the release workflow.
D:That corrective action is directed at a different model lifecycle management situation: A team records experiments but has no controlled path from a selected registered model to scheduled scoring jobs. The relevant corrective action here is add a backward-compatible schema migration and gated deployment order to the release workflow.