McAfee-Secured Website

Exam Code: AI-300

Exam Name: Operationalizing Machine Learning and Generative AI Solutions

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Machine Learning Operations Engineer Associate

100% Microsoft AI-300 Guaranteed Success With Testking Training Materials

Get Real AI-300 Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

213 Questions & Answers with Testing Engine

"Operationalizing Machine Learning and Generative AI Solutions Exam", also known as AI-300 exam, is a Microsoft certification exam.

Pass your tests with the always up-to-date AI-300 Exam Engine. Your AI-300 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Microsoft AI-300 Practice Test 16

Question 150

An experiment must compare an old production commit to a new branch without rewriting either. Which Git operation supports safe isolation?

  1. Require pull requests and code review before merging changes to the protected branch.
  2. Use small reviewed feature branches with merge conflict resolution against the current main branch.
  3. Create a separate comparison branch from the recorded production commit.
  4. Add appropriate generated-artifact patterns to the project's .gitignore.

Explanation

Correct Answer

C:

Independent branches preserve the production baseline while enabling controlled experiments and traceable diffs.

Incorrect Answers

A:

Require pull requests and code review before merging changes to the protected branch. That operation would be relevant if: A team pushes model training changes directly to the protected main branch and later cannot identify why production behavior changed. Its corresponding reason or verification is: Reviewed change history identifies the intent and author of modifications before those changes reach production automation. However the problem in this question is: An experiment must compare an old production commit to a new branch without rewriting either. The requirement instead calls for Create a separate comparison branch from the recorded production commit; the question-specific check is: Independent branches preserve the production baseline while enabling controlled experiments and traceable diffs.

B:

Use small reviewed feature branches with merge conflict resolution against the current main branch. A separate source case supports that approach when: Several engineers independently edit one YAML deployment configuration and merge conflicts repeatedly cause incorrect production parameters. The supporting rationale or observable result is: Controlled branch integration makes conflicting edits visible before production configuration is deployed. This question describes a different issue: An experiment must compare an old production commit to a new branch without rewriting either. Address it through Create a separate comparison branch from the recorded production commit, with the required justification or evidence: Independent branches preserve the production baseline while enabling controlled experiments and traceable diffs.

D:

Add appropriate generated-artifact patterns to the project's .gitignore. This is useful for the distinct issue: A model training script accidentally includes local experiment outputs and __pycache__ in each commit. The relevant technical rationale or evidence for that issue is: Ignoring reproducible caches and local outputs keeps source history focused on reviewed code and declarative configuration. Here the reported issue is: An experiment must compare an old production commit to a new branch without rewriting either. The relevant response is: Create a separate comparison branch from the recorded production commit. The first result cannot verify this different correction: Independent branches preserve the production baseline while enabling controlled experiments and traceable diffs.

Question 151

A production deployment must be rolled back after a bad merged code change. The organization wants to preserve the complete audit history. What approach should be used?

  1. Create a separate comparison branch from the recorded production commit.
  2. Revert the offending commit through a new reviewed change rather than rewriting main history.
  3. Use small reviewed feature branches with merge conflict resolution against the current main branch.
  4. Keep large datasets in governed object storage and version them as ML data assets.

Explanation

Correct Answer

B:

A revert records the corrective action transparently while leaving evidence of the original change intact.

Incorrect Answers

A:

The alternative, Create a separate comparison branch from the recorded production commit, belongs to a different technical condition: An experiment must compare an old production commit to a new branch without rewriting either. There the source-specific mechanism or proof is: Independent branches preserve the production baseline while enabling controlled experiments and traceable diffs. The condition asked about here is: A production deployment must be rolled back after a bad merged code change. The organization wants to preserve the complete audit history. A suitable correction is Revert the offending commit through a new reviewed change rather than rewriting main history; the targeted justification or result is: A revert records the corrective action transparently while leaving evidence of the original change intact.

C:

Use small reviewed feature branches with merge conflict resolution against the current main branch. That operation would be relevant if: Several engineers independently edit one YAML deployment configuration and merge conflicts repeatedly cause incorrect production parameters. Its corresponding reason or verification is: Controlled branch integration makes conflicting edits visible before production configuration is deployed. However the problem in this question is: A production deployment must be rolled back after a bad merged code change. The organization wants to preserve the complete audit history. The requirement instead calls for Revert the offending commit through a new reviewed change rather than rewriting main history; the question-specific check is: A revert records the corrective action transparently while leaving evidence of the original change intact.

D:

Keep large datasets in governed object storage and version them as ML data assets. A separate source case supports that approach when: A repository is becoming enormous because developers commit multi-gigabyte datasets beside Python source. The supporting rationale or observable result is: Git source control is not a replacement for scalable data versioning; separating data preserves manageable repository history. This question describes a different issue: A production deployment must be rolled back after a bad merged code change. The organization wants to preserve the complete audit history. Address it through Revert the offending commit through a new reviewed change rather than rewriting main history, with the required justification or evidence: A revert records the corrective action transparently while leaving evidence of the original change intact.

Question 152

An engineer commits personal notebook output containing customer rows while reviewing ML preprocessing logic. How should source-control rules be improved?

  1. Remove sensitive notebook outputs and enforce clean notebook/secret scanning before commits.
  2. Create a protected versioned release tag referencing the approved commit SHA.
  3. Revert the offending commit through a new reviewed change rather than rewriting main history.
  4. Use small reviewed feature branches with merge conflict resolution against the current main branch.

Explanation

Correct Answer

A:

Reviewable source files should not carry embedded private results; scanning and output cleaning reduce unintended data exposure.

Incorrect Answers

B:

Create a protected versioned release tag referencing the approved commit SHA. It addresses another case: An ML operations release must tie a deployed model back to exactly which source revision was approved. For that case, the distinguishing justification or acceptance check is: A tagged release provides an auditable marker that CI/CD can pin for deployments and future rollback. The present case instead states: An engineer commits personal notebook output containing customer rows while reviewing ML preprocessing logic. It calls for Remove sensitive notebook outputs and enforce clean notebook/secret scanning before commits. Showing success for the other case would not establish: Reviewable source files should not carry embedded private results; scanning and output cleaning reduce unintended data exposure.

C:

The alternative, Revert the offending commit through a new reviewed change rather than rewriting main history, belongs to a different technical condition: A production deployment must be rolled back after a bad merged code change. The organization wants to preserve the complete audit history. There the source-specific mechanism or proof is: A revert records the corrective action transparently while leaving evidence of the original change intact. The condition asked about here is: An engineer commits personal notebook output containing customer rows while reviewing ML preprocessing logic. A suitable correction is Remove sensitive notebook outputs and enforce clean notebook/secret scanning before commits; the targeted justification or result is: Reviewable source files should not carry embedded private results; scanning and output cleaning reduce unintended data exposure.

D:

Use small reviewed feature branches with merge conflict resolution against the current main branch. That operation would be relevant if: Several engineers independently edit one YAML deployment configuration and merge conflicts repeatedly cause incorrect production parameters. Its corresponding reason or verification is: Controlled branch integration makes conflicting edits visible before production configuration is deployed. However the problem in this question is: An engineer commits personal notebook output containing customer rows while reviewing ML preprocessing logic. The requirement instead calls for Remove sensitive notebook outputs and enforce clean notebook/secret scanning before commits; the question-specific check is: Reviewable source files should not carry embedded private results; scanning and output cleaning reduce unintended data exposure.

Question 153

Three training scripts log metrics under the Default MLflow experiment, making their project runs hard to compare. What should the author set before logging?

  1. Log the model artifact and relevant dataset or run metadata under the same tracked run.
  2. Verify MLflow tracking URI and authentication in the actual job execution environment.
  3. Log the confusion-matrix file as an MLflow run artifact.
  4. Set a meaningful MLflow experiment name for the related training runs.

Explanation

Correct Answer

D:

An experiment groups relevant runs and metrics under a stable project context without changing the underlying trained models.

Incorrect Answers

A:

Log the model artifact and relevant dataset or run metadata under the same tracked run. This is useful for the distinct issue: A production team wants all trained models and associated data lineage linked to the same run. The relevant technical rationale or evidence for that issue is: Consistent run linkage provides provenance between training inputs, metrics, and resulting model versions. Here the reported issue is: Three training scripts log metrics under the Default MLflow experiment, making their project runs hard to compare. The relevant response is: Set a meaningful MLflow experiment name for the related training runs. The first result cannot verify this different correction: An experiment groups relevant runs and metrics under a stable project context without changing the underlying trained models.

B:

Verify MLflow tracking URI and authentication in the actual job execution environment. It addresses another case: A scheduled notebook launches remote AML jobs, but its MLflow plots appear under a local tracking directory. For that case, the distinguishing justification or acceptance check is: Tracking configuration can differ between notebooks and containers; the remote process must connect to the intended tracking store. The present case instead states: Three training scripts log metrics under the Default MLflow experiment, making their project runs hard to compare. It calls for Set a meaningful MLflow experiment name for the related training runs. Showing success for the other case would not establish: An experiment groups relevant runs and metrics under a stable project context without changing the underlying trained models.

C:

The alternative, Log the confusion-matrix file as an MLflow run artifact, belongs to a different technical condition: A data scientist needs to preserve a confusion-matrix image generated during evaluation alongside an MLflow run. There the source-specific mechanism or proof is: Artifacts capture supplementary files such as plots and reports that are not scalar parameters or metrics. The condition asked about here is: Three training scripts log metrics under the Default MLflow experiment, making their project runs hard to compare. A suitable correction is Set a meaningful MLflow experiment name for the related training runs; the targeted justification or result is: An experiment groups relevant runs and metrics under a stable project context without changing the underlying trained models.

Question 154

An MLflow run stores a trained model but not its learning rate or regularization strength. Reviewers cannot reproduce the configuration. What should be logged?

  1. Configure MLflow to use the workspace's Azure Machine Learning tracking URI.
  2. Log the metric with a step index for every observed training epoch.
  3. Log the input hyperparameters using MLflow parameters for each run.
  4. Log the canonical release metric explicitly under a documented stable name.

Explanation

Correct Answer

C:

Parameters record the actual training configuration used so auditors can distinguish runs beyond the resulting model artifact.

Incorrect Answers

A:

Configure MLflow to use the workspace's Azure Machine Learning tracking URI. A separate source case supports that approach when: A team performs training from a laptop but wants MLflow runs visible in its Azure ML workspace. The supporting rationale or observable result is: The remote tracking endpoint associates externally launched runs with the intended workspace rather than an unrelated local MLflow store. This question describes a different issue: An MLflow run stores a trained model but not its learning rate or regularization strength. Reviewers cannot reproduce the configuration. Address it through Log the input hyperparameters using MLflow parameters for each run, with the required justification or evidence: Parameters record the actual training configuration used so auditors can distinguish runs beyond the resulting model artifact.

B:

Log the metric with a step index for every observed training epoch. This is useful for the distinct issue: A pipeline logged loss at every epoch, but reviewers see only its final value. The relevant technical rationale or evidence for that issue is: Step-indexed metrics preserve learning curves and help diagnose overfitting rather than reducing training to one scalar. Here the reported issue is: An MLflow run stores a trained model but not its learning rate or regularization strength. Reviewers cannot reproduce the configuration. The relevant response is: Log the input hyperparameters using MLflow parameters for each run. The first result cannot verify this different correction: Parameters record the actual training configuration used so auditors can distinguish runs beyond the resulting model artifact.

D:

Log the canonical release metric explicitly under a documented stable name. It addresses another case: Automated MLflow tracking is producing too many unexpected metric names, while explicit values are needed for a release gate. For that case, the distinguishing justification or acceptance check is: Release comparisons depend on stable metric keys; incidental autolog names can vary across frameworks and runs. The present case instead states: An MLflow run stores a trained model but not its learning rate or regularization strength. Reviewers cannot reproduce the configuration. It calls for Log the input hyperparameters using MLflow parameters for each run. Showing success for the other case would not establish: Parameters record the actual training configuration used so auditors can distinguish runs beyond the resulting model artifact.

Question 155

A training job prints its validation accuracy to stdout but the MLflow experiment comparison view shows no result. What action should be added?

  1. Log the confusion-matrix file as an MLflow run artifact.
  2. Log the validation accuracy using the MLflow metrics logging API.
  3. Log the canonical release metric explicitly under a documented stable name.
  4. Use MLflow's managed run context so termination status is recorded correctly.

Explanation

Correct Answer

B:

The tracking store requires explicit metric logging or supported autologging rather than relying on console output alone.

Incorrect Answers

A:

Log the confusion-matrix file as an MLflow run artifact. That operation would be relevant if: A data scientist needs to preserve a confusion-matrix image generated during evaluation alongside an MLflow run. Its corresponding reason or verification is: Artifacts capture supplementary files such as plots and reports that are not scalar parameters or metrics. However the problem in this question is: A training job prints its validation accuracy to stdout but the MLflow experiment comparison view shows no result. The requirement instead calls for Log the validation accuracy using the MLflow metrics logging API; the question-specific check is: The tracking store requires explicit metric logging or supported autologging rather than relying on console output alone.

C:

Log the canonical release metric explicitly under a documented stable name. A separate source case supports that approach when: Automated MLflow tracking is producing too many unexpected metric names, while explicit values are needed for a release gate. The supporting rationale or observable result is: Release comparisons depend on stable metric keys; incidental autolog names can vary across frameworks and runs. This question describes a different issue: A training job prints its validation accuracy to stdout but the MLflow experiment comparison view shows no result. Address it through Log the validation accuracy using the MLflow metrics logging API, with the required justification or evidence: The tracking store requires explicit metric logging or supported autologging rather than relying on console output alone.

D:

Use MLflow's managed run context so termination status is recorded correctly. This is useful for the distinct issue: A training script crashes after opening an MLflow run, leaving an unclear execution state. The relevant technical rationale or evidence for that issue is: Context management closes runs even when code exits unexpectedly, helping operators distinguish completed and failed experiments. Here the reported issue is: A training job prints its validation accuracy to stdout but the MLflow experiment comparison view shows no result. The relevant response is: Log the validation accuracy using the MLflow metrics logging API. The first result cannot verify this different correction: The tracking store requires explicit metric logging or supported autologging rather than relying on console output alone.

Question 156

A data scientist needs to preserve a confusion-matrix image generated during evaluation alongside an MLflow run. What should the script do?

  1. Log the confusion-matrix file as an MLflow run artifact.
  2. Configure MLflow to use the workspace's Azure Machine Learning tracking URI.
  3. Log the metric with a step index for every observed training epoch.
  4. Use MLflow's managed run context so termination status is recorded correctly.

Explanation

Correct Answer

A:

Artifacts capture supplementary files such as plots and reports that are not scalar parameters or metrics.

Incorrect Answers

B:

The alternative, Configure MLflow to use the workspace's Azure Machine Learning tracking URI, belongs to a different technical condition: A team performs training from a laptop but wants MLflow runs visible in its Azure ML workspace. There the source-specific mechanism or proof is: The remote tracking endpoint associates externally launched runs with the intended workspace rather than an unrelated local MLflow store. The condition asked about here is: A data scientist needs to preserve a confusion-matrix image generated during evaluation alongside an MLflow run. A suitable correction is Log the confusion-matrix file as an MLflow run artifact; the targeted justification or result is: Artifacts capture supplementary files such as plots and reports that are not scalar parameters or metrics.

C:

Log the metric with a step index for every observed training epoch. That operation would be relevant if: A pipeline logged loss at every epoch, but reviewers see only its final value. Its corresponding reason or verification is: Step-indexed metrics preserve learning curves and help diagnose overfitting rather than reducing training to one scalar. However the problem in this question is: A data scientist needs to preserve a confusion-matrix image generated during evaluation alongside an MLflow run. The requirement instead calls for Log the confusion-matrix file as an MLflow run artifact; the question-specific check is: Artifacts capture supplementary files such as plots and reports that are not scalar parameters or metrics.

D:

Use MLflow's managed run context so termination status is recorded correctly. A separate source case supports that approach when: A training script crashes after opening an MLflow run, leaving an unclear execution state. The supporting rationale or observable result is: Context management closes runs even when code exits unexpectedly, helping operators distinguish completed and failed experiments. This question describes a different issue: A data scientist needs to preserve a confusion-matrix image generated during evaluation alongside an MLflow run. Address it through Log the confusion-matrix file as an MLflow run artifact, with the required justification or evidence: Artifacts capture supplementary files such as plots and reports that are not scalar parameters or metrics.

Question 157

A team performs training from a laptop but wants MLflow runs visible in its Azure ML workspace. Where should the client send tracking data?

  1. Use MLflow's managed run context so termination status is recorded correctly.
  2. Log the metric with a step index for every observed training epoch.
  3. Verify MLflow tracking URI and authentication in the actual job execution environment.
  4. Configure MLflow to use the workspace's Azure Machine Learning tracking URI.

Explanation

Correct Answer

D:

The remote tracking endpoint associates externally launched runs with the intended workspace rather than an unrelated local MLflow store.

Incorrect Answers

A:

Use MLflow's managed run context so termination status is recorded correctly. It addresses another case: A training script crashes after opening an MLflow run, leaving an unclear execution state. For that case, the distinguishing justification or acceptance check is: Context management closes runs even when code exits unexpectedly, helping operators distinguish completed and failed experiments. The present case instead states: A team performs training from a laptop but wants MLflow runs visible in its Azure ML workspace. It calls for Configure MLflow to use the workspace's Azure Machine Learning tracking URI. Showing success for the other case would not establish: The remote tracking endpoint associates externally launched runs with the intended workspace rather than an unrelated local MLflow store.

B:

The alternative, Log the metric with a step index for every observed training epoch, belongs to a different technical condition: A pipeline logged loss at every epoch, but reviewers see only its final value. There the source-specific mechanism or proof is: Step-indexed metrics preserve learning curves and help diagnose overfitting rather than reducing training to one scalar. The condition asked about here is: A team performs training from a laptop but wants MLflow runs visible in its Azure ML workspace. A suitable correction is Configure MLflow to use the workspace's Azure Machine Learning tracking URI; the targeted justification or result is: The remote tracking endpoint associates externally launched runs with the intended workspace rather than an unrelated local MLflow store.

C:

Verify MLflow tracking URI and authentication in the actual job execution environment. That operation would be relevant if: A scheduled notebook launches remote AML jobs, but its MLflow plots appear under a local tracking directory. Its corresponding reason or verification is: Tracking configuration can differ between notebooks and containers; the remote process must connect to the intended tracking store. However the problem in this question is: A team performs training from a laptop but wants MLflow runs visible in its Azure ML workspace. The requirement instead calls for Configure MLflow to use the workspace's Azure Machine Learning tracking URI; the question-specific check is: The remote tracking endpoint associates externally launched runs with the intended workspace rather than an unrelated local MLflow store.

Question 158

A model's inference signature is missing, causing downstream clients to send unsupported column names. What tracking artifact should the team package?

  1. Log the metric with a step index for every observed training epoch.
  2. Log the canonical release metric explicitly under a documented stable name.
  3. Log or register the MLflow model with an input-output model signature.
  4. Log the input hyperparameters using MLflow parameters for each run.

Explanation

Correct Answer

C:

A model signature documents the expected schema and reduces integration mistakes between training and consuming applications.

Incorrect Answers

A:

Log the metric with a step index for every observed training epoch. This is useful for the distinct issue: A pipeline logged loss at every epoch, but reviewers see only its final value. The relevant technical rationale or evidence for that issue is: Step-indexed metrics preserve learning curves and help diagnose overfitting rather than reducing training to one scalar. Here the reported issue is: A model's inference signature is missing, causing downstream clients to send unsupported column names. The relevant response is: Log or register the MLflow model with an input-output model signature. The first result cannot verify this different correction: A model signature documents the expected schema and reduces integration mistakes between training and consuming applications.

B:

Log the canonical release metric explicitly under a documented stable name. It addresses another case: Automated MLflow tracking is producing too many unexpected metric names, while explicit values are needed for a release gate. For that case, the distinguishing justification or acceptance check is: Release comparisons depend on stable metric keys; incidental autolog names can vary across frameworks and runs. The present case instead states: A model's inference signature is missing, causing downstream clients to send unsupported column names. It calls for Log or register the MLflow model with an input-output model signature. Showing success for the other case would not establish: A model signature documents the expected schema and reduces integration mistakes between training and consuming applications.

D:

The alternative, Log the input hyperparameters using MLflow parameters for each run, belongs to a different technical condition: An MLflow run stores a trained model but not its learning rate or regularization strength. Reviewers cannot reproduce the configuration. There the source-specific mechanism or proof is: Parameters record the actual training configuration used so auditors can distinguish runs beyond the resulting model artifact. The condition asked about here is: A model's inference signature is missing, causing downstream clients to send unsupported column names. A suitable correction is Log or register the MLflow model with an input-output model signature; the targeted justification or result is: A model signature documents the expected schema and reduces integration mistakes between training and consuming applications.

Question 159

A training script crashes after opening an MLflow run, leaving an unclear execution state. What practice improves run lifecycle accuracy?

  1. Log the model artifact and relevant dataset or run metadata under the same tracked run.
  2. Use MLflow's managed run context so termination status is recorded correctly.
  3. Log the input hyperparameters using MLflow parameters for each run.
  4. Log the metric with a step index for every observed training epoch.

Explanation

Correct Answer

B:

Context management closes runs even when code exits unexpectedly, helping operators distinguish completed and failed experiments.

Incorrect Answers

A:

Log the model artifact and relevant dataset or run metadata under the same tracked run. A separate source case supports that approach when: A production team wants all trained models and associated data lineage linked to the same run. The supporting rationale or observable result is: Consistent run linkage provides provenance between training inputs, metrics, and resulting model versions. This question describes a different issue: A training script crashes after opening an MLflow run, leaving an unclear execution state. Address it through Use MLflow's managed run context so termination status is recorded correctly, with the required justification or evidence: Context management closes runs even when code exits unexpectedly, helping operators distinguish completed and failed experiments.

C:

Log the input hyperparameters using MLflow parameters for each run. This is useful for the distinct issue: An MLflow run stores a trained model but not its learning rate or regularization strength. Reviewers cannot reproduce the configuration. The relevant technical rationale or evidence for that issue is: Parameters record the actual training configuration used so auditors can distinguish runs beyond the resulting model artifact. Here the reported issue is: A training script crashes after opening an MLflow run, leaving an unclear execution state. The relevant response is: Use MLflow's managed run context so termination status is recorded correctly. The first result cannot verify this different correction: Context management closes runs even when code exits unexpectedly, helping operators distinguish completed and failed experiments.

D:

Log the metric with a step index for every observed training epoch. It addresses another case: A pipeline logged loss at every epoch, but reviewers see only its final value. For that case, the distinguishing justification or acceptance check is: Step-indexed metrics preserve learning curves and help diagnose overfitting rather than reducing training to one scalar. The present case instead states: A training script crashes after opening an MLflow run, leaving an unclear execution state. It calls for Use MLflow's managed run context so termination status is recorded correctly. Showing success for the other case would not establish: Context management closes runs even when code exits unexpectedly, helping operators distinguish completed and failed experiments.

Question 160

A production team wants all trained models and associated data lineage linked to the same run. What MLflow logging practice helps?

  1. Log the model artifact and relevant dataset or run metadata under the same tracked run.
  2. Log or register the MLflow model with an input-output model signature.
  3. Configure MLflow to use the workspace's Azure Machine Learning tracking URI.
  4. Log the confusion-matrix file as an MLflow run artifact.

Explanation

Correct Answer

A:

Consistent run linkage provides provenance between training inputs, metrics, and resulting model versions.

Incorrect Answers

B:

Log or register the MLflow model with an input-output model signature. A separate source case supports that approach when: A model's inference signature is missing, causing downstream clients to send unsupported column names. The supporting rationale or observable result is: A model signature documents the expected schema and reduces integration mistakes between training and consuming applications. This question describes a different issue: A production team wants all trained models and associated data lineage linked to the same run. Address it through Log the model artifact and relevant dataset or run metadata under the same tracked run, with the required justification or evidence: Consistent run linkage provides provenance between training inputs, metrics, and resulting model versions.

C:

Configure MLflow to use the workspace's Azure Machine Learning tracking URI. This is useful for the distinct issue: A team performs training from a laptop but wants MLflow runs visible in its Azure ML workspace. The relevant technical rationale or evidence for that issue is: The remote tracking endpoint associates externally launched runs with the intended workspace rather than an unrelated local MLflow store. Here the reported issue is: A production team wants all trained models and associated data lineage linked to the same run. The relevant response is: Log the model artifact and relevant dataset or run metadata under the same tracked run. The first result cannot verify this different correction: Consistent run linkage provides provenance between training inputs, metrics, and resulting model versions.

D:

Log the confusion-matrix file as an MLflow run artifact. It addresses another case: A data scientist needs to preserve a confusion-matrix image generated during evaluation alongside an MLflow run. For that case, the distinguishing justification or acceptance check is: Artifacts capture supplementary files such as plots and reports that are not scalar parameters or metrics. The present case instead states: A production team wants all trained models and associated data lineage linked to the same run. It calls for Log the model artifact and relevant dataset or run metadata under the same tracked run. Showing success for the other case would not establish: Consistent run linkage provides provenance between training inputs, metrics, and resulting model versions.

Operationalizing Machine Learning and Generative AI Solutions (AI-300) Practice Tests