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 2

Question 10

Consider the following fictional Databricks production case: A single-node model training run spends almost all time swapping a 40 GB feature matrix on a 16 GB VM; no distributed version exists. Which implementation is most appropriate for the stated incident?

  1. Increase the training host's memory within budget or change to a supported distributed estimator.
  2. Assign each trial a unique tracked run with a shared parent experiment and stable trial metadata.
  3. Adopt validated model-parallel or sharded training to fit the model and its optimizer state.
  4. Prepare a versioned reusable feature snapshot and schedule Ray trials against appropriately shared data.

Explanation

Correct Answer

A:

This corrects the relevant failure boundary instead of merely hiding the symptom. Increase the training host's memory within budget or change to a supported distributed estimator. The issue matters because training stalls on memory pressure even while unrelated cluster nodes are underutilized.

Incorrect Answers

B:

That corrective action is directed at a different scaling and tuning situation: An Optuna optimization reuses the same MLflow run ID for concurrent trials, causing overlapping parameter values and artifacts. The relevant corrective action here is increase the training host's memory within budget or change to a supported distributed estimator.

C:

That corrective action is directed at a different scaling and tuning situation: A language model's weights cannot fit on one accelerator yet a team launches pure data-parallel replicas, each loading the entire model. The relevant corrective action here is increase the training host's memory within budget or change to a supported distributed estimator.

D:

That corrective action 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 relevant corrective action here is increase the training host's memory within budget or change to a supported distributed estimator.

Question 11

Review this production model incident: A model performance dashboard compares this week's labels to last week's predictions without joining by stable request ID or consistent outcome horizon. Which failure risk follows most closely from the unresolved behavior?

  1. The live population drifts far from model training even though daily changes remain modest.
  2. Early-account fraud receives inadequate detection despite acceptable combined recall.
  3. The metric fails or silently excludes malformed observations, distorting the reported trend.
  4. Errors are assigned to the wrong predictions and reported performance becomes meaningless.

Explanation

Correct Answer

D:

This consequence follows from the specific unresolved condition. Errors are assigned to the wrong predictions and reported performance becomes meaningless. The causal constraint is: An inference table needs correct prediction-to-label linkage for trustworthy observed performance metrics.

Incorrect Answers

A:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: A model trained on last year's customer mix is monitored only against yesterday's already shifted cohort, so its original training-population mismatch is hidden. The more immediate risk here is errors are assigned to the wrong predictions and reported performance becomes meaningless.

B:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: An account-risk model's fraud recall is stable overall but unexpectedly falls for new customers in their first 30 days. The more immediate risk here is errors are assigned to the wrong predictions and reported performance becomes meaningless.

C:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: An ML owner adds a custom metric computed from the average of a string-typed currency column without validating conversion. The more immediate risk here is errors are assigned to the wrong predictions and reported performance becomes meaningless.

Question 12

In an enterprise ML workflow, the reported issue is: A model is promoted to production with its code unchanged but a different unpinned training library version is installed, altering prediction preprocessing. What likely operational or model-quality harm does this issue create?

  1. Apparently identical code yields different results or fails to load under the new runtime.
  2. The artifact can disappear with an account change or be overwritten without reliable provenance.
  3. The wrong experiment artifact may be used in production because lifecycle stages are disconnected.
  4. Post-incident investigators cannot recreate the prediction or explain the changed model behavior.

Explanation

Correct Answer

A:

This consequence follows from the specific unresolved condition. Apparently identical code yields different results or fails to load under the new runtime. The causal constraint is: Environment transitions must preserve or explicitly test library and runtime dependencies as well as source code.

Incorrect Answers

B:

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 apparently identical code yields different results or fails to load under the new runtime.

C:

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 apparently identical code yields different results or fails to load under the new runtime.

D:

That direct consequence 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 more immediate risk here is apparently identical code yields different results or fails to load under the new runtime.

Question 13

In an enterprise ML workflow, the reported issue is: A retailer needs one independent forecasting model per store but concatenates all store histories before fitting one global model. What technical distinction is essential to this investigation?

  1. Group-specific training must preserve group identity when independent model parameters are required.
  2. Horizontal replica count alone does not combine GPU memory for one model without a supported parallelization strategy.
  3. Ray actor/task orchestration can fit fine-grained stateful workloads that do not map naturally to relational Spark queries.
  4. Parallelizable fraction and upstream service limits bound the benefits of adding model-training workers.

Explanation

Correct Answer

A:

The technical constraint is specific to the reported failure. Group-specific training must preserve group identity when independent model parameters are required. To test it, compare the business requirement and per-store holdout behavior with the single global estimator's outputs.

Incorrect Answers

B:

That underlying principle 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 current incident instead turns on group-specific training must preserve group identity when independent model parameters are required.

C:

That underlying principle is directed at a different scaling and tuning situation: A reinforcement-learning simulator requires stateful actors exchanging messages and irregular task graphs, but the team plans one Spark SQL groupBy per simulation step. The current incident instead turns on group-specific training must preserve group identity when independent model parameters are required.

D:

That underlying principle is directed at a different scaling and tuning situation: An ML project has a single sequential external feature-service dependency and expects data-parallel workers to increase throughput 20-fold. The current incident instead turns on group-specific training must preserve group identity when independent model parameters are required.

Question 14

Review this production model incident: A monitoring team wants a business metric for the share of transactions exceeding a risk threshold but records only the raw count of high-risk transactions. Which mitigation resolves the defect rather than concealing it?

  1. Tune the proven bottleneck and concurrency/capacity settings instead of scaling CPU blindly.
  2. Add an appropriately sustained alert condition or window aggregate instead of a one-minute peak trigger.
  3. Define a documented custom rate metric with safe zero-denominator handling in profiling outputs.
  4. Require approved aggregation, privacy safeguards and sample sufficiency for sensitive slices.

Explanation

Correct Answer

C:

This corrects the relevant failure boundary instead of merely hiding the symptom. Define a documented custom rate metric with safe zero-denominator handling in profiling outputs. The issue matters because traffic growth is mistaken for increasing risk because the raw count rises mechanically.

Incorrect Answers

A:

That corrective action is directed at a different drift detection and lakehouse monitoring situation: A production endpoint scales CPU instances up when traffic rises, but request queues and P95 response time keep growing despite low CPU usage. The relevant corrective action here is define a documented custom rate metric with safe zero-denominator handling in profiling outputs.

B:

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 a documented custom rate metric with safe zero-denominator handling in profiling outputs.

D:

That corrective action is directed at a different drift detection and lakehouse monitoring situation: A demographic fairness review uses a tiny sensitive subgroup with three labeled examples and reports its observed error as definitive. The relevant corrective action here is define a documented custom rate metric with safe zero-denominator handling in profiling outputs.

Question 15

In an enterprise ML workflow, the reported issue is: A fraud PyFunc uses a global mutable feature cache initialized by the first request and mixes data between subsequent users. Before modifying the pipeline, which evidence should be collected first?

  1. Replay interleaved synthetic customer requests and inspect whether cache keys and outputs remain isolated.
  2. Inspect the run's metric keys versus artifact store and recover the expected plot file.
  3. Correlate worker logs and metric histories by run ID to expose simultaneous writes to the same run.
  4. Inspect parent_run_id tags, candidate run IDs and the experiment's tracked search configuration.

Explanation

Correct Answer

A:

This is the most discriminating check for the observed symptoms. Replay interleaved synthetic customer requests and inspect whether cache keys and outputs remain isolated. The underlying technical issue is that request-scoped feature retrieval and thread-safe state are needed when the custom model serves multiple callers.

Incorrect Answers

B:

That diagnostic evidence is directed at a different advanced mlflow usage situation: An ML engineer logs a calibration plot as a string metric, then downstream reviewers cannot inspect the visualization or its input version. For the stated incident, the decisive observation is replay interleaved synthetic customer requests and inspect whether cache keys and outputs remain isolated.

C:

That diagnostic evidence is directed at a different advanced mlflow usage situation: A training orchestrator starts parallel child fits but all workers accidentally log into one active MLflow run and overwrite each other's metrics. For the stated incident, the decisive observation is replay interleaved synthetic customer requests and inspect whether cache keys and outputs remain isolated.

D:

That diagnostic evidence is directed at a different advanced mlflow usage situation: An experiment tunes 40 candidates but logs each trial as an unrelated root MLflow run, so reviewers cannot reconstruct which search produced a winner. For the stated incident, the decisive observation is replay interleaved synthetic customer requests and inspect whether cache keys and outputs remain isolated.

Question 16

Review this production model incident: A model's inference table records output probabilities but omits ground-truth labels, yet reviewers expect immediate precision and recall metrics. Which failure risk follows most closely from the unresolved behavior?

  1. Investigators cannot tell whether a particular model version produced the anomalous scores.
  2. Material drift remains unaddressed because nobody receives the notification.
  3. Configuration fails on nonexistent inference fields and delays basic table monitoring.
  4. A dashboard displays unsupported performance claims based only on predicted probabilities.

Explanation

Correct Answer

D:

This consequence follows from the specific unresolved condition. A dashboard displays unsupported performance claims based only on predicted probabilities. The causal constraint is: An inference profile can monitor inputs and predictions without labels, but true predictive performance requires observed outcomes.

Incorrect Answers

A:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: A serving endpoint appears healthy but nobody records model version, request identifier or prediction value, so a drift incident cannot be reconstructed. The more immediate risk here is a dashboard displays unsupported performance claims based only on predicted probabilities.

B:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: A fraud-feature drift monitor crosses its threshold but the alert has no assigned recipient and no incident ownership. The more immediate risk here is a dashboard displays unsupported performance claims based only on predicted probabilities.

C:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: An analyst wants to inspect today’s static customer table for missing values and profile changes but chooses an inference monitor requiring model IDs and predictions. The more immediate risk here is a dashboard displays unsupported performance claims based only on predicted probabilities.

Question 17

Consider the following fictional Databricks production case: An ML engineer tries to parallelize a stateful neural-training loop by executing independent mapPartitions calls that share one mutable global model in the driver. What consequence most specifically follows from leaving the issue unchanged?

  1. The rewritten job runs only on one worker or fails during model serialization.
  2. A larger cluster runs the hyperparameter search slower because each task is overhead-dominated.
  3. Worker updates diverge because the assumed global shared model does not exist.
  4. Unnecessary model partitioning adds network latency with little memory benefit.

Explanation

Correct Answer

C:

This consequence follows from the specific unresolved condition. Worker updates diverge because the assumed global shared model does not exist. The causal constraint is: Separate Spark tasks do not safely share or synchronously mutate one Python model instance across workers.

Incorrect Answers

A:

That direct consequence is directed at a different scaling and tuning situation: A team switches a SparkML training stage to Ray without verifying that its chosen estimator supports Ray's distributed fit contracts. The more immediate risk here is worker updates diverge because the assumed global shared model does not exist.

B:

That direct consequence is directed at a different scaling and tuning situation: A researcher distributes 500 tiny one-second trials across many Ray workers and sees coordination cost exceed compute time. The more immediate risk here is worker updates diverge because the assumed global shared model does not exist.

D:

That direct consequence is directed at a different scaling and tuning situation: A tabular classifier fits comfortably on one GPU and 100 million independent examples are available, but a team proposes splitting the small model's layers across hosts. The more immediate risk here is worker updates diverge because the assumed global shared model does not exist.

Question 18

In an enterprise ML workflow, the reported issue is: A baseline table is overwritten every night with the current day's data before a drift calculation, making the comparison always look unchanged. What likely operational or model-quality harm does this issue create?

  1. The cold-start cohort receives poor recommendations while the combined metric remains acceptable.
  2. Real drift is erased because the comparator silently changes with the observed sample.
  3. A small but important region remains unavailable while overall service metrics seem healthy.
  4. Concept drift or changed label relationships degrade decisions without an input-drift alarm.

Explanation

Correct Answer

B:

This consequence follows from the specific unresolved condition. Real drift is erased because the comparator silently changes with the observed sample. The causal constraint is: A fixed baseline must be versioned and immutable for the intended comparison period.

Incorrect Answers

A:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: A recommender reports stable overall accuracy while click-through prediction quality for first-time users deteriorates sharply. The more immediate risk here is real drift is erased because the comparator silently changes with the observed sample.

C:

That direct consequence is directed at a different drift detection and lakehouse monitoring situation: An alert query uses one global transaction-failure count and masks a complete outage in a small high-value region. The more immediate risk here is real drift is erased because the comparator silently changes with the observed sample.

D:

That direct consequence 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. The more immediate risk here is real drift is erased because the comparator silently changes with the observed sample.

Question 19

The engineering team encounters the following issue: A churn feature job succeeds and a classifier achieves acceptable F1, but the deployed endpoint returns HTTP 400 because request column names differ. What corrective action should the engineer prefer?

  1. Build an end-to-end integration fixture that verifies schemas, versions and prediction outputs across stages.
  2. Guard zero-variance inputs and assert expected fallback or explicit failure semantics in unit tests.
  3. Add an environment-level integration test for table, model and serving input/output contracts.
  4. Keep reusable logic and tests in separate versioned units with explicit CI test stages.

Explanation

Correct Answer

A:

This corrects the relevant failure boundary instead of merely hiding the symptom. Build an end-to-end integration fixture that verifies schemas, versions and prediction outputs across stages. The issue matters because a model that looks accurate offline cannot be called successfully by production clients.

Incorrect Answers

B:

That corrective action is directed at a different validation testing situation: A preprocessing function divides by a standard deviation and returns infinity for a constant numeric column. The relevant corrective action here is build an end-to-end integration fixture that verifies schemas, versions and prediction outputs across stages.

C:

That corrective action is directed at a different validation testing situation: A team unit-tests each model component independently but its feature table and serving endpoint disagree on column names. The relevant corrective action here is build an end-to-end integration fixture that verifies schemas, versions and prediction outputs across stages.

D:

That corrective action 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. The relevant corrective action here is build an end-to-end integration fixture that verifies schemas, versions and prediction outputs across stages.

Question 20

A production ML review contains this situation: A deployed custom model fetches latest feature values during serving while offline evaluation used event-time snapshots without checking the difference. What is the most material immediate risk in this situation?

  1. Online performance degrades because the model receives different kinds of feature values than it learned.
  2. Duplicate or omitted trials distort comparison when two partial histories are treated as independent full searches.
  3. Important diagnostic evidence is absent from model review despite a reported scalar score.
  4. Tracking artifacts create unauthorized exposure paths for customers' personal data.

Explanation

Correct Answer

A:

This consequence follows from the specific unresolved condition. Online performance degrades because the model receives different kinds of feature values than it learned. The causal constraint is: The model's request-time feature contract should be compared with training-time feature semantics to avoid unintended skew.

Incorrect Answers

B:

That direct consequence is directed at a different advanced mlflow usage situation: A research team restarts a long hyperparameter sweep and merges all newly executed trials into a new experiment without connecting them to the original parent. The more immediate risk here is online performance degrades because the model receives different kinds of feature values than it learned.

C:

That direct consequence is directed at a different advanced mlflow usage situation: An ML engineer logs a calibration plot as a string metric, then downstream reviewers cannot inspect the visualization or its input version. The more immediate risk here is online performance degrades because the model receives different kinds of feature values than it learned.

D:

That direct consequence is directed at a different advanced mlflow usage situation: A regulated analytics team uploads full customer identifiers inside MLflow model artifacts during troubleshooting. The more immediate risk here is online performance degrades because the model receives different kinds of feature values than it learned.

Certified Machine Learning Professional Practice Tests