Oracle Cloud Infrastructure Data Science Professional (2024): Hands-On Workshop — Lesson 2: ADS Data Access, Exploration, Cleaning, Feature Engineering, and Dataset Versioning

Learning Objectives

• Interpret the architecture, roles, data, rules, dependencies, and expected business result for ads data access, exploration, cleaning, feature engineering, and dataset versioning.
• Perform the required configuration, development, reporting, administration, or transaction workflow in a controlled environment.
• Prove the outcome using project configuration, notebook environments, data profiles, feature transformations, experiment runs, evaluation metrics, explanation output, model metadata, endpoint tests, and monitoring logs and explain remaining operational risk.

Detailed Topics

• ADS Data Access
• Exploration
• Cleaning
• Feature Engineering
• Dataset Versioning
• requirements, ownership, prerequisites, data, and acceptance criteria for ads data access, exploration, cleaning, feature engineering, and dataset versioning
• security, controls, integration, performance, recovery, and support implications for ads data access, exploration, cleaning, feature engineering, and dataset versioning

Instructor Demonstration

Demonstrate ads data access, exploration, cleaning, feature engineering, and dataset versioning within a governed machine-learning solution built in OCI Data Science from reproducible data preparation through evaluated deployment and monitored inference. Use representative data and capture project configuration, notebook environments, data profiles, feature transformations, experiment runs, evaluation metrics, explanation output, model metadata, endpoint tests, and monitoring logs so another practitioner can independently verify the result.

Hands-On Exercise

Complete a controlled Oracle Cloud Infrastructure Data Science Professional (2024): Hands-On Workshop exercise focused on ads data access, exploration, cleaning, feature engineering, and dataset versioning. Introduce one realistic failure involving unreproducible environments, data leakage, poor feature quality, unsuitable metrics, biased models, missing metadata, insecure endpoints, drift, and failed retraining; isolate the cause, apply the smallest reversible correction, and rerun the acceptance checks.

1. Record the requirement, environment, identities, dependencies, data, controls, and expected result.
2. Perform the lesson workflow and capture evidence before and after each material change.
3. Test normal, boundary, failure, security, reconciliation, recovery, and operational conditions relevant to the topic.
4. Document the root cause, correction, rollback, owner, support action, and next improvement priority.

Scenario Analysis

• Identify the process owner, technical owner, approver, affected users, and support team.
• Compare the likelihood and impact of unreproducible environments, data leakage, poor feature quality, unsuitable metrics, biased models, missing metadata, insecure endpoints, drift, and failed retraining against the proposed design or control.
• Define escalation, rollback, communications, acceptance evidence, and post-change monitoring.

Troubleshooting Guidance

When ads data access, exploration, cleaning, feature engineering, and dataset versioning does not produce the expected result, confirm scope, effective dates, identities, data context, dependencies, and recent changes. Review the Oracle-specific logs, reports, traces, calculations, workflow history, transaction status, or diagnostic output relevant to the lesson. Test one reversible hypothesis at a time and verify that the correction does not create unreproducible environments, data leakage, poor feature quality, unsuitable metrics, biased models, missing metadata, insecure endpoints, drift, and failed retraining elsewhere in the process.

Validation Checklist

• Prerequisites, ownership, data context, settings, and expected outcomes are documented.
• The workflow is reproducible and supported by project configuration, notebook environments, data profiles, feature transformations, experiment runs, evaluation metrics, explanation output, model metadata, endpoint tests, and monitoring logs.
• Security, controls, data quality, integration, performance, recovery, and support conditions are reviewed.
• Exceptions, corrective action, rollback, approvals, monitoring, and handover are recorded.

Lesson Outcome

Learners can apply ADS Data Access, Exploration, Cleaning, Feature Engineering, and Dataset Versioning in a realistic Oracle Cloud Infrastructure Data Science Professional (2024): Hands-On Workshop situation and defend the result with course-specific evidence.