cert
cert-1
cert-2

Pass Databricks Certified Machine Learning Professional Exam in First Attempt Guaranteed!

Get 100% Latest Exam Questions, Accurate & Verified Answers to Pass the Actual Exam!
30 Days Free Updates, Instant Download!

cert-5
cert-6
Certified Machine Learning Professional Exam - Verified By Experts
Certified Machine Learning Professional Premium File

Certified Machine Learning Professional Premium File

$69.99
$76.99
  • Premium File 216 Questions & Answers. Last Update: Nov 12, 2026

Whats Included:

  • Latest Questions
  • 100% Accurate Answers
  • Fast Exam Updates
 
$76.99
$69.99
accept 12 downloads in the last 7 days
block-screenshots
Databricks Certified Machine Learning Professional Practice Test 75

Question 740

In an enterprise ML workflow, the reported issue is: A global model predicts late deliveries accurately overall, but all mistakes occur at one small coastal depot and the team never slices its results by site. Before modifying the pipeline, which evidence should be collected first?

  1. Inspect inference table version IDs and label-matched errors before and after traffic shift.
  2. Compare model errors and sample sizes by depot against the aggregate baseline.
  3. Compare per-replica resident memory and allocation limits before and after model promotion.
  4. Stratify historical samples by day type and examine whether the observed difference is expected seasonality.

Explanation

Correct Answer

B:

This is the most discriminating check for the observed symptoms. Compare model errors and sample sizes by depot against the aggregate baseline. The underlying technical issue is that slicing on operationally meaningful locations can reveal concentrated failures hidden in the global metric.

Incorrect Answers

A:

That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A scoring endpoint quietly moves from model v7 to v8 during a rollout but an inference trend aggregates both versions in one error curve. For the stated incident, the decisive observation is compare model errors and sample sizes by depot against the aggregate baseline.

C:

That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: A new model version uses twice the memory per replica, but capacity monitoring tracks only CPU and crash counts. For the stated incident, the decisive observation is compare model errors and sample sizes by depot against the aggregate baseline.

D:

That diagnostic evidence is directed at a different drift detection and lakehouse monitoring situation: An online feature monitor compares weekend shopping to weekday reference data and treats expected weekend behavior as a model failure. For the stated incident, the decisive observation is compare model errors and sample sizes by depot against the aggregate baseline.

Question 741

In an enterprise ML workflow, the reported issue is: A travel site serves 30,000 bookings per minute and wants to evaluate a new fraud model while limiting customer impact during its first hour. Before modifying the pipeline, which evidence should be collected first?

  1. Replay stored valid requests from old and new clients against the candidate endpoint before shifting traffic.
  2. Compare serving-error, latency and outcome metrics separately for champion and canary model versions at the proposed traffic share.
  3. Compare effective serving traffic split, model identities and alias resolution in the live endpoint.
  4. Inspect endpoint state, individual served-entity readiness and model-load errors after the update.

Explanation

Correct Answer

B:

This is the most discriminating check for the observed symptoms. Compare serving-error, latency and outcome metrics separately for champion and canary model versions at the proposed traffic share. The underlying technical issue is that a canary rollout can expose a small controlled share of live traffic while retaining a proven model for most requests.

Incorrect Answers

A:

That diagnostic evidence is directed at a different deployment strategies situation: A rollout updates the serving model but a renamed input column breaks requests from an older mobile client. For the stated incident, the decisive observation is compare serving-error, latency and outcome metrics separately for champion and canary model versions at the proposed traffic share.

C:

That diagnostic evidence is directed at a different deployment strategies situation: During a controlled release, the candidate model's P95 latency triples and engineers want to change the registered model alias without adjusting endpoint routing. For the stated incident, the decisive observation is compare serving-error, latency and outcome metrics separately for champion and canary model versions at the proposed traffic share.

D:

That diagnostic evidence is directed at a different deployment strategies situation: An endpoint update request succeeds but the newly configured model version is still loading; monitoring marks the release complete immediately. For the stated incident, the decisive observation is compare serving-error, latency and outcome metrics separately for champion and canary model versions at the proposed traffic share.

Question 742

Review this production model incident: A fraud detector predicts the majority no-fraud class for 99.4% of cases and advertises 99.4% accuracy while missing most positives. Which mitigation resolves the defect rather than concealing it?

  1. Hold out a final test set before tuning and score the selected pipeline exactly once after selection.
  2. Deploy a governed low-latency serving endpoint with compatible feature retrieval and load-test capacity.
  3. Evaluate the model with class-sensitive metrics and choose thresholds aligned with false-positive and missed-fraud tradeoffs.
  4. Apply VectorAssembler to produce the features vector before fitting the chosen learning algorithm.

Explanation

Correct Answer

C:

This corrects the relevant failure boundary instead of merely hiding the symptom. Evaluate the model with class-sensitive metrics and choose thresholds aligned with false-positive and missed-fraud tradeoffs. The issue matters because important fraud cases remain undetected despite an impressive accuracy percentage.

Incorrect Answers

A:

That corrective action is directed at a different using spark ml situation: After choosing the best parameter map, a data scientist reports the same cross-validation score as an independent final test result without any untouched data. The relevant corrective action here is evaluate the model with class-sensitive metrics and choose thresholds aligned with false-positive and missed-fraud tradeoffs.

B:

That corrective action is directed at a different using spark ml situation: An online pricing service requires a personalized response within 70 milliseconds per request but its team schedules a full overnight batch job as the only inference route. The relevant corrective action here is evaluate the model with class-sensitive metrics and choose thresholds aligned with false-positive and missed-fraud tradeoffs.

D:

That corrective action is directed at a different using spark ml situation: A team wants to convert several numeric inputs into one features vector but treats VectorAssembler as a clustering Estimator that should output predictions. The relevant corrective action here is evaluate the model with class-sensitive metrics and choose thresholds aligned with false-positive and missed-fraud tradeoffs.

Question 743

Consider the following fictional Databricks production case: A hyperparameter team compares two Optuna studies with different train/validation splits and chooses the lower loss as though both metrics were directly comparable. What underlying principle determines the most appropriate interpretation?

  1. Combining Spark feature preparation with Ray trial execution requires explicit materialization and data locality planning.
  2. Data-parallel workloads may benefit from more executor cores and appropriately partitioned tasks rather than vertical driver scaling.
  3. Independent per-group models are suitable for distributing grouped pandas-function work, subject to available compute and partition size.
  4. Tuning runs must use equivalent evaluation data and metric definitions before ranking configurations.

Explanation

Correct Answer

D:

The technical constraint is specific to the reported failure. Tuning runs must use equivalent evaluation data and metric definitions before ranking configurations. To test it, compare dataset version, random seed, fold definition and metric contract attached to each MLflow run.

Incorrect Answers

A:

That underlying principle is directed at a different scaling and tuning situation: A hyperparameter search operates on large Spark-managed feature tables but each candidate model is small and independent, and workers repeatedly convert the entire table for each Ray task. The current incident instead turns on tuning runs must use equivalent evaluation data and metric definitions before ranking configurations.

B:

That underlying principle is directed at a different scaling and tuning situation: A feature-engineering job is CPU-bound on hundreds of independent daily partitions and a team upgrades only the driver to a much larger machine. The current incident instead turns on tuning runs must use equivalent evaluation data and metric definitions before ranking configurations.

C:

That underlying principle is directed at a different scaling and tuning situation: An ecommerce analyst fits one independent scikit-learn model per 3,000 product categories, but the driver sequentially loops through all category subsets. The current incident instead turns on tuning runs must use equivalent evaluation data and metric definitions before ranking configurations.

Question 744

A production ML review contains this situation: A production endpoint scales CPU instances up when traffic rises, but request queues and P95 response time keep growing despite low CPU usage. What production-safe change best matches the evidence?

  1. Join labels to their originating predictions using stable IDs and a defined maturity window.
  2. Normalize timestamp semantics before applying approved local-hour performance slices.
  3. Add an appropriately sustained alert condition or window aggregate instead of a one-minute peak trigger.
  4. Tune the proven bottleneck and concurrency/capacity settings instead of scaling CPU blindly.

Explanation

Correct Answer

D:

This corrects the relevant failure boundary instead of merely hiding the symptom. Tune the proven bottleneck and concurrency/capacity settings instead of scaling CPU blindly. The issue matters because the deployment wastes compute while inference delays and timeouts continue increasing.

Incorrect Answers

A:

That corrective action is directed at a different drift detection and lakehouse monitoring situation: A model performance dashboard compares this week's labels to last week's predictions without joining by stable request ID or consistent outcome horizon. The relevant corrective action here is tune the proven bottleneck and concurrency/capacity settings instead of scaling CPU blindly.

B:

That corrective action is directed at a different drift detection and lakehouse monitoring situation: A model's prediction error appears to spike at 2 a.m. local time but the monitoring table groups timestamps in UTC for a globally distributed region. The relevant corrective action here is tune the proven bottleneck and concurrency/capacity settings instead of scaling CPU blindly.

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 tune the proven bottleneck and concurrency/capacity settings instead of scaling CPU blindly.

Question 745

Consider the following fictional Databricks production case: A training dataset combines a feature table with multiple timestamped rows per device using an equality join on device ID only. Which result is the strongest acceptance criterion after remediation?

  1. A fixture with three feature revisions produces exactly one eligible feature row per prediction time.
  2. A failover replay produces the same feature totals as an uninterrupted control run.
  3. Renaming a display label does not alter the entity's online lookup result.
  4. Equivalent test events produce matching approved feature values across both paths.

Explanation

Correct Answer

A:

This verification measures the required result rather than just process liveness. A fixture with three feature revisions produces exactly one eligible feature row per prediction time. It confirms whether the targeted action succeeded: configure time-series columns and a timestamp lookup in the supported feature training-set API.

Incorrect Answers

B:

That acceptance test is directed at a different advanced feature store concepts situation: An online behavioral feature counter consumes a streaming source but does not retain checkpoint state after worker restarts. Here the actual success criterion is a fixture with three feature revisions produces exactly one eligible feature row per prediction time.

C:

That acceptance test is directed at a different advanced feature store concepts situation: A low-latency feature table uses external customer names as lookup keys; names change and are not unique. Here the actual success criterion is a fixture with three feature revisions produces exactly one eligible feature row per prediction time.

D:

That acceptance test 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. Here the actual success criterion is a fixture with three feature revisions produces exactly one eligible feature row per prediction time.

Question 746

A production ML review contains this situation: A churn experiment separately scripts feature scaling in training but omits that scaling when the deployed prediction job loads the classifier. Which check shows that this particular problem has been resolved?

  1. A held-out review confirms alert volume and true-positive capture at the frozen threshold.
  2. Two sample sentences produce the expected token-array column from a direct transform.
  3. A stored raw fixture produces matching transformed feature vectors and predictions in training and deployment.
  4. Both known and unknown fixture categories yield the documented safe scoring behavior.

Explanation

Correct Answer

C:

This verification measures the required result rather than just process liveness. A stored raw fixture produces matching transformed feature vectors and predictions in training and deployment. It confirms whether the targeted action succeeded: fit and persist a unified PipelineModel containing scaling and model stages, then use it for inference.

Incorrect Answers

A:

That acceptance test is directed at a different using spark ml situation: A binary classifier ranks risky cases well but sends too many alerts at a default 0.5 decision threshold. Here the actual success criterion is a stored raw fixture produces matching transformed feature vectors and predictions in training and deployment.

B:

That acceptance test is directed at a different using spark ml situation: A text pipeline repeatedly fits a Tokenizer stage even though tokenization simply converts strings into token arrays without learning a model. Here the actual success criterion is a stored raw fixture produces matching transformed feature vectors and predictions in training and deployment.

D:

That acceptance test is directed at a different using spark ml situation: An inference pipeline sends new categorical labels unseen during training to StringIndexerModel with error handling configured to reject unknown categories. Here the actual success criterion is a stored raw fixture produces matching transformed feature vectors and predictions in training and deployment.

Question 747

A production ML review contains this situation: A new recommendation model consumes three times as much memory and the team plans a canary based solely on prediction accuracy, ignoring serving capacity. What production-safe change best matches the evidence?

  1. Add a version-compatible request adapter or coordinate client migration with explicit acceptance gates.
  2. Gate release completion on ready healthy serving entities and validated predictions.
  3. Set and verify the intended approved candidate-versus-champion traffic split before exposure.
  4. Require resource and response-time acceptance criteria before expanding canary traffic.

Explanation

Correct Answer

D:

This corrects the relevant failure boundary instead of merely hiding the symptom. Require resource and response-time acceptance criteria before expanding canary traffic. The issue matters because the new version overwhelms replicas and creates customer-facing timeouts despite good offline accuracy.

Incorrect Answers

A:

That corrective action is directed at a different deployment strategies situation: A rollout updates the serving model but a renamed input column breaks requests from an older mobile client. The relevant corrective action here is require resource and response-time acceptance criteria before expanding canary traffic.

B:

That corrective action is directed at a different deployment strategies situation: An endpoint update request succeeds but the newly configured model version is still loading; monitoring marks the release complete immediately. The relevant corrective action here is require resource and response-time acceptance criteria before expanding canary traffic.

C:

That corrective action is directed at a different deployment strategies situation: A Databricks Model Serving endpoint hosts champion and candidate versions, but a release config accidentally routes 100% of traffic to the candidate instead of 10%. The relevant corrective action here is require resource and response-time acceptance criteria before expanding canary traffic.

Question 748

Review this production model incident: An engineer uses only production replay data to validate a code change and exposes customer events in an unrestricted test workspace. What evidence would best distinguish the actual fault from neighboring ML concerns?

  1. Inspect test dataset sensitivity, workspace access rights and synthetic alternative fixtures.
  2. Inspect registered model versions, alias resolution and active serving configuration after the release.
  3. Inspect notebook entrypoints, test invocation conditions and deployment job definitions.
  4. Run a tiny constant-column fixture and inspect finite-value validation after transformation.

Explanation

Correct Answer

A:

This is the most discriminating check for the observed symptoms. Inspect test dataset sensitivity, workspace access rights and synthetic alternative fixtures. The underlying technical issue is that test stages need appropriate data classification and isolation, not just increasingly realistic inputs.

Incorrect Answers

B:

That diagnostic evidence is directed at a different validation testing situation: A model version is registered but the deployment job points to an old alias after promotion and continues serving last week's model. For the stated incident, the decisive observation is inspect test dataset sensitivity, workspace access rights and synthetic alternative fixtures.

C:

That diagnostic evidence is directed at a different validation testing situation: An ML project stores test files directly in each production notebook and execution of a deployed notebook inadvertently reruns expensive tests. For the stated incident, the decisive observation is inspect test dataset sensitivity, workspace access rights and synthetic alternative fixtures.

D:

That diagnostic evidence is directed at a different validation testing situation: A preprocessing function divides by a standard deviation and returns infinity for a constant numeric column. For the stated incident, the decisive observation is inspect test dataset sensitivity, workspace access rights and synthetic alternative fixtures.

Question 749

Review this production model incident: A training dataset combines a feature table with multiple timestamped rows per device using an equality join on device ID only. Which failure risk follows most closely from the unresolved behavior?

  1. Training labels duplicate and features from future measurements become attached to older events.
  2. Late sensor readings are assigned to the wrong time window and mislead incident models.
  3. Several models change their effective input distributions without deliberate retraining or review.
  4. A repeated refresh doubles feature rows and multiplies downstream training joins.

Explanation

Correct Answer

A:

This consequence follows from the specific unresolved condition. Training labels duplicate and features from future measurements become attached to older events. The causal constraint is: A time-series feature table needs a documented timestamp key and an as-of lookup to avoid multiplying labels or attaching future values.

Incorrect Answers

B:

That direct consequence is directed at a different advanced feature store concepts situation: A sensor feature pipeline computes rolling 30-minute averages by arrival time, although delayed events carry trustworthy event timestamps. The more immediate risk here is training labels duplicate and features from future measurements become attached to older events.

C:

That direct consequence 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 more immediate risk here is training labels duplicate and features from future measurements become attached to older events.

D:

That direct consequence 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 more immediate risk here is training labels duplicate and features from future measurements become attached to older events.

Question 750

In an enterprise ML workflow, the reported issue is: A model release requires all clients from a restricted region to keep the old version, but a simple percentage split sends some of those clients to the candidate. What likely operational or model-quality harm does this issue create?

  1. The team cannot recover from a bad promotion without an emergency rebuild.
  2. All requests are exposed to the candidate during what operators believed was a small canary.
  3. Regulated traffic reaches an unapproved model version contrary to residency controls.
  4. Older clients receive validation errors even though new client requests succeed.

Explanation

Correct Answer

C:

This consequence follows from the specific unresolved condition. Regulated traffic reaches an unapproved model version contrary to residency controls. The causal constraint is: Random traffic splitting alone does not guarantee policy-based tenant or regional routing.

Incorrect Answers

A:

That direct consequence is directed at a different deployment strategies situation: An ML operator promotes a new model version but its automated rollback points to a model that was deleted from the endpoint configuration. The more immediate risk here is regulated traffic reaches an unapproved model version contrary to residency controls.

B:

That direct consequence is directed at a different deployment strategies situation: A Databricks Model Serving endpoint hosts champion and candidate versions, but a release config accidentally routes 100% of traffic to the candidate instead of 10%. The more immediate risk here is regulated traffic reaches an unapproved model version contrary to residency controls.

D:

That direct consequence is directed at a different deployment strategies situation: A rollout updates the serving model but a renamed input column breaks requests from an older mobile client. The more immediate risk here is regulated traffic reaches an unapproved model version contrary to residency controls.

Certified Machine Learning Professional Practice Tests