Detailed Answer in Step-by-Step Solution:

Define Feature Engineering: It’s the process of creating or modifying features to improve model performance.

Evaluate Options:

A: Parameter tuning adjusts model hyperparameters (e.g., learning rate), not features.

B: Model interpretation (e.g., SHAP values) explains predictions, not feature creation.

C: Transforming features (e.g., normalizing, encoding) is the core of feature engineering—correct.

D: Understanding features occurs during exploration, not engineering.

Reasoning: Feature engineering directly manipulates data inputs (e.g., converting timestamps to day-of-week), distinct from tuning or interpretation.

Conclusion: C is the precise definition.

OCI Data Science documentation defines feature engineering as “the process of transforming raw data into features that better represent the underlying problem to the predictive models, resulting in improved model accuracy.” Examples include scaling or creating interaction terms, aligning with C. Other options (A, B, D) relate to different ML stages.

Oracle Cloud Infrastructure Data Science Documentation, "Feature Engineering Overview".

Detailed Answer in Step-by-Step Solution:

Objective: Identify the non-visualized AutoML stage with small data.

Understand AutoML Pipeline: Includes sampling, feature/algorithm selection, tuning.

Evaluate Options:

A: Feature selection—Visualized (e.g., feature importance).

B: Algorithm selection—Visualized (e.g., algorithm scores).

C: Adaptive sampling—Skipped/visualization absent for <1000 rows.

D: Hyperparameter tuning—Visualized (e.g., trial plots).

Reasoning: Adaptive sampling optimizes large datasets; small data skips it, omitting visuals.

Conclusion: C is correct.

OCI AutoML documentation notes: “Adaptive sampling is applied to large datasets (>1000 rows) to reduce size; for smaller datasets, it’s skipped, and no visualization is generated.” Other stages (A, B,D) produce visuals—only C is absent here.

Oracle Cloud Infrastructure AutoML Documentation, "Pipeline Stages".

Detailed Answer in Step-by-Step Solution:

Objective: Find a true statement about OCI Anomaly Detection.

Understand Service: Detects anomalies in multivariate data (e.g., time series).

Evaluate Options:

A: False—Accepted types are CSV/JSON, not SQL/Python.

B: Partially true—Focuses on numerical data (e.g., sensors), not text broadly.

C: True—Used for fraud, intrusions, and sensor anomalies (key use cases).

D: False—Trained on customer data only, not general datasets.

Reasoning: C aligns with documented applications; others misalign.

Conclusion: C is correct.

OCI Anomaly Detection documentation states: “The service is designed to detect anomalies in time series data, making it valuable for fraud detection, network intrusion analysis, and sensor discrepancies.” A is incorrect (file formats), B overgeneralizes (numerical focus), and D misstates training data—only C matches the service’s purpose.

Oracle Cloud Infrastructure Anomaly Detection Documentation, "Use Cases".

Detailed Answer in Step-by-Step Solution:

Objective: Identify a widely accepted maxim about data scientists’ time allocation.

Understand Data Science Workflow: Involves data collection, preparation, and analysis—time distribution is key.

Evaluate Options:

A: 80% on finding/preparing, 20% analyzing—Reflects the data wrangling challenge.

B: 80% analyzing, 20% finding/preparing—Inverts the common perception.

C: 80% on failed projects, 20% useful—Pessimistic, not a standard maxim.

Reasoning: Industry consensus (e.g., “80/20 rule”) emphasizes data prep as the bulk of effort due to messy real-world data.

Conclusion: A is correct.

OCI Data Science documentation aligns with industry norms: “Data scientists typically spend 80% of their time finding, cleaning, and preparing data, and 20% on analysis and modeling, due to the complexity of raw data.” B reverses this, and C isn’t supported—only A reflects this widely cited maxim from sources like Forbes and OCI’s practical guidance.

Oracle Cloud Infrastructure Data Science Documentation, "Data Science WorkflowOverview".

Detailed Answer in Step-by-Step Solution:

Objective: Identify which activity isn’t part of the ML lifecycle.

Define ML Lifecycle: Includes data access, preparation, modeling, evaluation, deployment, and monitoring.

Evaluate Options:

A: Database Management (e.g., DBA tasks) is IT-related, not specific to ML workflows.

B: Model Deployment (e.g., serving predictions) is a key ML phase—correctly included.

C: Modeling (e.g., training) is the core of ML—correctly included.

D: Data Access (e.g., retrieving data) is the first ML step—correctly included.

Reasoning: Database management supports infrastructure, not the ML process directly.

Conclusion: A is the outlier.

The OCI Data Science lifecycle includes “data access, exploration, feature engineering, modeling, deployment, and monitoring,” per the documentation. Database Management (A) is a general ITtask (e.g., optimizing Oracle DB), not an ML-specific activity, unlike B, C, and D, which are integral to OCI’s ML pipeline.

Oracle Cloud Infrastructure Data Science Documentation, "Machine Learning Lifecycle Overview".

Detailed Answer in Step-by-Step Solution:

Objective: Calculate optimal bandwidth for model deployment.

Given Data:

Payload size = 1024 KB = 1024 * 8 = 8192 Kb (kilobits).

Requests/sec = 120.

Buffer = 20% = 0.2.

Calculate Base Bandwidth:

Bits/sec = Payload * Requests = 8192 Kb * 120 = 983,040 Kb/s = 983.04 Mbps.

Add Buffer:

Total = Base * (1 + Buffer) = 983.04 * 1.2 = 1179.648 Mbps.

Evaluate Options: Closest to 1179.648 Mbps is 1152 Mbps (D)—realistic rounding.

Conclusion: D is correct.

OCI documentation advises: “Calculate bandwidth as payload size (in bits) * requests/sec, then add a buffer (e.g., 20%) for peak loads.” Here, 1024 KB = 8192 Kb, * 120 = 983.04 Mbps, * 1.2 = 1179.648 Mbps. D (1152 Mbps) is the closest practical option—452 (A) and 52 (B) are too low, 7052 (C) excessive.

Oracle Cloud Infrastructure Data Science Documentation, "Model Deployment - Load Balancing".

Detailed Answer in Step-by-Step Solution:

Objective: Identify supported annotation classes in OCI Data Labeling.

Understand Tool: Supports image/text annotations for ML.

Evaluate Options:

A: Object detection—Yes (bounding boxes).

B: Named entity—Text-specific, not primary for images.

C: Classification—Yes (labels for images/text).

D: Key-point—Not listed in OCI docs.

E: Polygonal—Not explicitly supported.

F: Semantic segmentation—Yes (pixel-level).

Reasoning: A, C, F match OCI’s image/text focus.

Conclusion: A, C, F are correct.

OCI Data Labeling supports “object detection (A), classification (C), and semantic segmentation (F) for images and text,” per documentation. B is text-specific, D and E aren’t highlighted—only A, C, F are core classes.

Oracle Cloud Infrastructure Data Labeling Documentation, "Annotation Types".

Detailed Answer in Step-by-Step Solution:

Objective: Diagnose a network timeout during pip install in a notebook session.

Understand Notebook Networking: Sessions run in a VCN; internet access requires specific configs.

Analyze Timeout: Indicates failure to reach the public PyPI repository—likely no internet outbound route.

Evaluate Options:

A: NAT Gateway—Provides internet access for private subnets—correct fix.

B: Service Gateway—Accesses OCI services privately, not public internet.

C: FastConnect—Links to on-premises, not public internet.

D: VNIC—Essential but present by default; doesn’t solve internet access.

Reasoning: NAT Gateway enables outbound traffic to public repos like PyPI.

Conclusion: A is correct.

OCI documentation notes: “Notebook sessions in a private subnet require a NAT Gateway to access public internet resources, such as PyPI, for package installation via pip. Without it, network timeouts occur.” Service Gateway (B) is for OCI services, FastConnect (C) is irrelevant, and VNIC (D) is standard—only A resolves the issue.

Oracle Cloud Infrastructure Data Science Documentation, "Networking for Notebook Sessions".

Detailed Answer in Step-by-Step Solution:

Objective: Identify the OCI Data Science interaction method that doesn’t require scripting.

Understand Interaction Methods: OCI provides multiple ways to interact with Data Science services—some are GUI-based, others script-based.

Evaluate Options:

A. OCI Console: A web-based graphical interface allowing users to manage resources (e.g., create notebook sessions, deploy models) via point-and-click—no scripting needed.

B. CLI: Command Line Interface requires writing commands (scripts) to execute tasks (e.g., oci data-science notebook-session create).

C. Language SDKs: Software Development Kits (e.g., Python SDK) require coding to interact programmatically (e.g., oci.data_science.DataScienceClient).

D. REST APIs: Application Programming Interfaces require scripted HTTP requests (e.g., using curl or a programming language).

Reasoning: Only the OCI Console (A) offers a no-code, user-friendly interface, while B, C, and D rely on scripting or programming.

Conclusion: A is the correct answer as it eliminates the need for scripting.

The OCI Console is described in the documentation as “a browser-based interface that allows users to manage OCI Data Science resources, such as creating notebook sessions or jobs, without writing code or scripts.” In contrast, the CLI (B) requires command-line scripts, SDKs (C) need programming (e.g., Python), and REST APIs (D) involve scripted API calls. The Console’s GUI distinguishes it as the only option functioning without scripting, aligning with Oracle’s design for accessibility to non-programmers.

Oracle Cloud Infrastructure Data Science Documentation, "Getting Started with OCI Console" section.

Detailed Answer in Step-by-Step Solution:

Objective: Understand FAP’s effect in OCI Anomaly Detection.

Evaluate Options:

A: Disable reporting—Incorrect; FAP sets threshold.

B: Sensitivity—Correct; lower FAP reduces false positives.

C: Error message—Incorrect; not a count mechanism.

D: Score per signal—Incorrect; FAP is a global setting.

Reasoning: FAP adjusts detection threshold—key to sensitivity.

Conclusion: B is correct.

OCI documentation states: “False Alarm Probability (FAP) controls the model’s sensitivity—lower values reduce false positives, higher values increase detection.” B aligns—others misrepresent FAP’s role.

Oracle Cloud Infrastructure Anomaly Detection Documentation, "FAP Configuration".

Detailed Answer in Step-by-Step Solution:

Objective: Identify the non-Observability and Management (O&M) service in OCI.

Understand O&M: Includes monitoring, logging, events tools.

Evaluate Options:

A: Event Services—Triggers actions, part of O&M—correct.

B: OCI Management Service—Not a defined O&M service—incorrect.

C: Logging Analytics—Log analysis, O&M component—correct.

D: Logging—Log collection, O&M component—correct.

Reasoning: B isn’t listed in OCI’s O&M suite—others are.

Conclusion: B is correct (not part of O&M).

OCI documentation lists “Observability and Management Services as including Event Services (A), Logging Analytics (C), and Logging (D)—‘OCI Management Service’ (B) is not a recognized component.” B appears to be a misnomer—only A, C, D are O&M per OCI’s service catalog.

Oracle Cloud Infrastructure Observability and Management Documentation, "Service Overview".

Detailed Answer in Step-by-Step Solution:

Objective: Identify ADS class for accessing datasets (e.g., scikit-learn).

Evaluate Options:

A: DatasetBrowser—Not an ADS class.

B: DatasetFactory—Loads datasets from sources like scikit-learn—correct.

C: ADSTuner—Hyperparameter tuning, not data access.

D: SecretKeeper—Manages credentials, not datasets.

Reasoning: DatasetFactory simplifies dataset loading (e.g., DatasetFactory.open()).

Conclusion: B is correct.

OCI documentation states: “DatasetFactory in ADS SDK provides methods to easily load datasets from libraries like scikit-learn or other sources (e.g., DatasetFactory.open('sklearn.datasets:load_iris')).” A isn’t real, C tunes models, and D handles secrets—only B fits.

Oracle Cloud Infrastructure ADS SDK Documentation, "DatasetFactory".

Detailed Answer in Step-by-Step Solution:

Objective: Identify the Oracle service to sort document images by language.

Task Breakdown: Requires extracting text from images (OCR) and detecting language—two potential services involved.

Evaluate Options:

A. Oracle Digital Assistant: Builds chatbots—irrelevant to image or language processing.

B. OCI Language: Detects and classifies languages in text—ideal for sorting after text extraction.

C. OCI Speech: Transcribes audio to text—not applicable to images.

D. OCI Vision: Performs OCR to extract text from images—necessary but not sufficient for language sorting.

Reasoning: The task emphasizes “sorting by language.” OCI Vision extracts text, but OCI Language identifies the language (e.g., English, Spanish). Since the question asks for one service and focuses on sorting, OCI Language (B) is the best fit, assuming text extraction is a precursor step.

Conclusion: B is correct.

OCI Language “provides language detection and classification capabilities, enabling identification of languages in text extracted from documents,” per the documentation. OCI Vision handles OCR, but the sorting task aligns with OCI Language (B). Digital Assistant (A) and Speech (C) don’t apply, and while Vision (D) is a prerequisite, B is the primary service for language sorting as per OCI’s AI service design.

Oracle Cloud Infrastructure Language Documentation, "Language Detection Features".

Detailed Answer in Step-by-Step Solution:

Objective: Locate stdout/stderr output from score.py in deployment.

Understand Deployment: score.py runs in a model endpoint; logs are managed by OCI.

Evaluate Options:

A: False—No VM file is defined by default; logs go to OCI Logging.

B: False—Console displays UI, not raw logs.

C: False—Cloud Shell is a separate tool, not a log destination.

D: True—Predict logs in OCI Logging capture stdout/stderr.

Reasoning: OCI centralizes logs in its Logging service for deployments.

Conclusion: D is correct.

OCI documentation states: “During model deployment, stdout and stderr from score.py are automatically sent to the predict log in the OCI Logging service, as configured in the deployment settings.” A, B, and C don’t align with this managed logging approach—only D is accurate.

Oracle Cloud Infrastructure Data Science Documentation, "Model Deployment - Logging".

Detailed Answer in Step-by-Step Solution:

Objective: Classify a PagerDuty notification in OCI Monitoring.

Understand OCI Monitoring: Involves events, rules, and actions for notifications.

Evaluate Options:

A: Action—executes a response (e.g., notify PagerDuty) when triggered—correct.

B: Rule—defines conditions for triggering actions—precedes the action.

C: Function—serverless code, not directly tied to notifications.

D: Event—state change triggering a rule, not the notification itself.

Reasoning: The notification is the action taken after an event/rule—fits A.

Conclusion: A is correct.

OCI documentation states: “Actions in the Monitoring service execute responses, such as sending notifications to PagerDuty, when a rule’s condition is met based on an event.” Rules (B) set conditions, Functions (C) are unrelated, and Events (D) are triggers—only Action (A) describes the notification step.

Oracle Cloud Infrastructure Monitoring Documentation, "Actions Overview".

Detailed Answer in Step-by-Step Solution:

Objective: Identify two OCI Language AI capabilities.

Understand OCI Language: Focuses on text analysis tasks.

Evaluate Options:

A: Table extraction—Vision, not Language—incorrect.

B: Punctuation correction—Not offered—incorrect.

C: Sentence diagramming—Not supported—incorrect.

D: Topic classification—Supported (custom/pretrained)—correct.

E: Sentiment analysis—Supported (pretrained)—correct.

Reasoning: D and E are core text analysis features of OCI Language.

Conclusion: D and E are correct.

OCI documentation states: “OCI Language offers topic classification (D) and sentiment analysis (E) for text analysis, among other features.” A belongs to Vision, B and C aren’t available—only D and E match OCI Language’s capabilities.

Oracle Cloud Infrastructure Language Documentation, "Text Analysis Features".

Detailed Answer in Step-by-Step Solution:

Objective: Identify export formats for OCI Data Labeling.

Understand Export: Annotated data is exported for ML use.

Evaluate Options:

A: CONLL V2003—Text-specific, not OCI-supported.

B: COCO—Standard for image annotations—correct.

C: Proprietary JSON—OCI’s default format—correct.

D: Spacy—NLP format, not supported.

Reasoning: B and C are image-compatible per OCI docs.

Conclusion: B and C are correct.

OCI documentation states: “Data Labeling exports annotations in COCO format (B) for image tasks and a proprietary JSON format (C) specific to the service.” CONLL (A) and Spacy (D) are text/NLP-focused—not supported for OCI’s image annotations.

Oracle Cloud Infrastructure Data Labeling Documentation, "Export Formats".

Detailed Answer in Step-by-Step Solution:

Objective: Identify a true statement about OCI Jobs logging.

Understand Logging: Jobs can log stdout/stderr to OCI Logging service.

Evaluate Options:

A: False—Each run has its own log, not a single job log.

B: False—Logging is optional, not mandatory.

C: True—When enabled, stdout/stderr are auto-captured.

D: False—Logs persist unless explicitly deleted.

Reasoning: C matches OCI’s automatic logging feature.

Conclusion: C is correct.

OCI documentation states: “When automatic log creation is enabled for Data Science Jobs, all stdout and stderr outputs are captured and stored in the OCI Logging service.” A is incorrect (per-run logs), B is optional, and D contradicts log retention—only C is accurate.

Oracle Cloud Infrastructure Data Science Documentation, "Jobs Logging".

Detailed Answer in Step-by-Step Solution:

Objective: Select an ADSTuner strategy to minimize cost and assess model quality.

Understand ADSTuner: Optimizes hyperparameters with configurable search spaces.

Evaluate Options:

A: Detailed—Exhaustive, high cost—incorrect.

B: No search space—False; tuning requires a space.

C: Perfunctory—Quick, low-cost assessment—correct.

D: Dictionary—Defines space but not a strategy.

Reasoning: Perfunctory balances cost and initial quality check.

Conclusion: C is correct.

OCI documentation states: “ADSTuner’s perfunctory strategy (C) performs a quick, low-cost search to assess model quality, ideal for initial tuning.” Detailed (A) is costly, B misstates requirements, and D is a method, not a strategy—only C fits the goal.

Oracle Cloud Infrastructure ADS SDK Documentation, "ADSTuner Search Strategies".

Detailed Answer in Step-by-Step Solution:

Objective: Efficiently manage varying hyperparameters in OCI Data Science Jobs.

Understand OCI Jobs: Jobs execute predefined tasks with configurable inputs (e.g., env vars, args).

Evaluate Options:

A: New job per run with env vars—Redundant job creation, inefficient.

B: New job per run with args—Similarly inefficient due to repeated setup.

C: Hardcode params, new job per change—Highly inefficient, requires code edits.

D: Single job, flexible params via env vars or args—Efficient, reusable—correct.

Reasoning: D minimizes job creation, allows runtime flexibility via configuration—best practice.

Conclusion: D is correct.

OCI documentation states: “For Jobs with varying hyperparameters, write code to accept environment variables or command-line arguments (D), then configure these per Job Run using the OCI Console or SDK—most efficient approach.” Options A, B, and C involve unnecessary job proliferation or code changes—only D aligns with OCI’s design for parameterized runs.

Oracle Cloud Infrastructure Data Science Documentation, "Configuring Job Runs with Parameters".

Detailed Answer in Step-by-Step Solution:

Objective: Identify the OCI service for scalable Spark applications.

Evaluate Options:

A: Data Science—ML platform, not Spark-focused.

B: Anomaly Detection—Specific ML service, not general Spark.

C: Data Labeling—Annotation tool, not Spark-related.

D: Data Flow—Managed Spark service for big data.

Reasoning: Data Flow is OCI’s Spark execution engine.

Conclusion: D is correct.

OCI Data Flow “provides a fully managed environment to run Apache Spark applications at scale, ideal for data processing and ML tasks.” Data Science (A) supports Spark in notebooks, but Data Flow (D) is the dedicated, scalable solution—B and C are unrelated.

Oracle Cloud Infrastructure Data Flow Documentation, "Overview".

Detailed Answer in Step-by-Step Solution:

Objective: Select an OCI AI service for text and table extraction from documents.

Evaluate Options:

A: Language—Text analysis, not extraction—incorrect.

B: Digital Assistant—Chatbots, not document tasks—incorrect.

C: Speech—Audio transcription, not documents—incorrect.

D: Vision—OCR for text/tables—correct.

Reasoning: Vision’s OCR extracts text and tables from document images.

Conclusion: D is correct.

OCI documentation states: “OCI Vision (D) uses OCR to extract text and tables from documents, supporting document analysis tasks.” A analyzes text post-extraction, B and C are unrelated—only D fits per OCI’s AI services.

Oracle Cloud Infrastructure Vision Documentation, "Document Analysis Features".

Detailed Answer in Step-by-Step Solution:

Understand OCI Data Labeling Capabilities: OCI Data Labeling is designed to annotate data for machine learning, supporting specific file types like images, text documents, and videos.

Evaluate Options:

A. Video footage: Supported for tasks like object detection or action recognition.

B. Images: Supported for image classification, object detection, etc.

C. Typewritten document: Supported as text data for tasks like entity extraction or classification.

D. Purchase orders: While potentially text-based, this is ambiguous without format clarification (e.g., PDF, image). OCI supports text annotation, but “purchase orders” isn’t a specific file type—it’s assumed as text here.

E. Audio recording: Not supported, as OCI Data Labeling focuses on visual and textual data, not audio.

Select Three: A (video), B (images), and C (text documents) are explicitly supported file types.

OCI Data Labeling supports annotating datasets of images, text, and videos, as per the official documentation. Video footage (A) can be annotated for tasks like object tracking, images (B) for classification or detection, and typewritten documents (C) for text-based annotations (e.g., named entity recognition). Audio files (E) are not supported, and while purchase orders (D) could be text, the question specifies “typewritten document” as a clearer match. (Reference: Oracle Cloud Infrastructure Data Labeling Service Documentation, "Supported Data Types").

Detailed Answer in Step-by-Step Solution:

Understand ORM Template: It automates OCI Data Science setup with predefined configurations.

Evaluate Components:

A: User groups are created for role-based access—automated.

B: Dynamic groups (e.g., for notebook sessions) are included—automated.

C: Individual users require manual creation via IAM—not automated.

D: Basic policies (e.g., access to Data Science resources) are included—automated.

Reasoning: ORM focuses on infrastructure and permissions, not user accounts.

Conclusion: C is the exception.

The OCI Resource Manager template for Data Science “automatically provisions user groups, dynamic groups, and policies for basic use cases,” but “individual users must be created separately in IAM and assigned to groups.” C is the only item not handled by the template, per the documentation.

Oracle Cloud Infrastructure Resource Manager Documentation, "Data Science Template".

Detailed Answer in Step-by-Step Solution:

Objective: Encode categorical features in a Data Science context (likely ADS SDK).

Understand Encoding: Converts categories (e.g., countries) to numerical forms.

Evaluate Options:

A: Not a standard ADS method—incorrect.

B: General transformation, not specific encoding—incorrect.

C: OneHotEncoder—Standard for categorical encoding—correct.

D: Visualization, not encoding—incorrect.

Reasoning: One-hot encoding creates binary columns—ideal for multiple categories.

Conclusion: C is correct.

OCI documentation states: “In ADS SDK, use OneHotEncoder (C) from sklearn (or similar) to encode categorical features like ‘countries’ into binary vectors for modeling.” A isn’t real, B is too broad, D is unrelated—only C fits OCI’s encoding practice.

Oracle Cloud Infrastructure Data Science Documentation, "Feature Encoding with ADS".

Detailed Answer in Step-by-Step Solution:

Objective: Efficiently increase storage for an OCI Job.

Understand Jobs: Storage (block volume) is set at job creation, not dynamically adjustable.

Evaluate Options:

A: False—Jobs can’t edit storage post-creation; it’s fixed.

B: False—No environment variable adjusts storage size.

C: True—Create a new job with larger storage (e.g., 200 GB) and run it.

D: False—Refactoring code is inefficient compared to increasing storage.

Reasoning: C is the standard OCI process for adjusting resources.

Conclusion: C is correct.

OCI documentation states: “Storage size for a Data Science Job is specified during job creation (e.g., block volume size). To increase it, create a new job with a larger storage configuration and initiate a new run.” Editing (A) isn’t supported, variables (B) don’t apply, and refactoring (D) avoids the issue—only C is efficient.

Oracle Cloud Infrastructure Data Science Documentation, "Jobs - Storage Configuration".

Detailed Answer in Step-by-Step Solution:

Understand OCI Anomaly Detection: This service trains models to detect anomalies in multivariate data (e.g., sensor readings), requiring data to be accessible within OCI’s ecosystem.

Assess Storage Requirements: The training data must be in a scalable, OCI-compatible location that the Anomaly Detection service can access programmatically.

Evaluate Options:

A. Your local machine: Data on a local machine isn’t directly accessible to OCI services without upload, making it impractical for cloud-based training.

B. MySQL database: While OCI supports MySQL, Anomaly Detection doesn’t natively integrate with it for training data; it prefers file-based input.

C. Autonomous Data Warehouse: This is a database for analytics, not the default storage for Anomaly Detection training data, which expects CSV/JSON files.

D. Object Storage Bucket: OCI Object Storage is a scalable, durable storage service that Anomaly Detection uses to ingest training data (e.g., CSV files).

Reasoning: Object Storage is the standard for large-scale data in OCI services, offering seamless integration with Anomaly Detection via APIs or SDKs.

Conclusion: D is the correct choice as it aligns with the service’s architecture.

The OCI Anomaly Detection service requires training data to be uploaded to an Object Storage Bucket in formats like CSV or JSON. This is explicitly outlined in the official documentation, which states that users must “upload the training dataset to an OCI Object Storage bucket” before creating a data asset for model training. Local storage (A) isn’t viable for cloud processing, and databases like MySQL (B) or Autonomous Data Warehouse (C) aren’t supported as primary inputs. Object Storage (D) provides the scalability and accessibility needed for multivariate anomaly detection workflows.

Oracle Cloud Infrastructure Anomaly Detection Documentation, "Preparing Training Data" section.

Detailed Answer in Step-by-Step Solution:

Objective: Identify the OCI service for ML model lifecycle.

Evaluate Options:

A: Data Catalog—Metadata management, not ML.

B: Data Integration—ETL, not ML.

C: Data Science—Full ML lifecycle—correct.

D: Data Flow—Spark processing, not full ML.

Reasoning: C supports building, training, deploying models.

Conclusion: C is correct.

OCI documentation states: “OCI Data Science (C) provides tools to build, train, and deploy machine learning models in the cloud, including notebooks and model catalog.” A, B, and D serve other purposes—only C fits the ML lifecycle per OCI’s offerings.

Oracle Cloud Infrastructure Data Science Documentation, "Service Overview".

Detailed Answer in Step-by-Step Solution:

Objective: Identify protection for data at rest/transit with customer control.

Evaluate Options:

A: Controls—Broad, not specific to encryption.

B: Isolation—Separates tenants, not crypto-focused.

C: Encryption—Secures data, allows key management—correct.

D: Federation—Auth sharing, not data protection.

Reasoning: C provides crypto control (e.g., Vault keys).

Conclusion: C is correct.

OCI documentation states: “Data encryption (C) protects data at rest and in transit, with customer-managed keys in OCI Vault meeting compliance needs.” A and B are broader, D is unrelated—only C fits per OCI’s security model.

Oracle Cloud Infrastructure Security Documentation, "Data Encryption".

Detailed Answer in Step-by-Step Solution:

Objective: Specify third-party libraries for model deployment.

Understand Artifacts: runtime.yaml defines runtime; score.py handles logic.

Evaluate Options:

A: Not a standard file—incorrect.

B: Inference code—not for dependencies.

C: Defines conda env with dependencies—correct.

D: Pip list—not used in OCI conda deployments.

Reasoning: runtime.yaml points to a conda env with all libraries.

Conclusion: C is correct.

OCI documentation states: “In runtime.yaml, specify the conda environment slug (e.g., ENVIRONMENT_SLUG: custom_env) containing all third-party libraries required by the model.” score.py (B) is for logic, requirements.txt (D) isn’t OCI-standard, and A doesn’t exist—C fixes the issue.

Oracle Cloud Infrastructure Data Science Documentation, "Model Deployment - runtime.yaml".

Detailed Answer in Step-by-Step Solution:

Objective: Determine where to specify the conda environment for an OCI model deployment.

Understand Model Deployment: Requires artifacts like score.py and runtime.yaml to define runtime settings.

Evaluate Options:

A. score.py: Contains inference logic (e.g., load_model(), predict())—not for environment specs.

B. runtime.yaml: Defines deployment runtime, including conda environment path—correct.

C. requirements.txt: Lists pip dependencies—not used in OCI for conda environments.

D. model_artifact_validate.py: Not a standard artifact; doesn’t exist in OCI deployment.

Reasoning: runtime.yaml specifies the conda env (e.g., slug: pyspark30_p37_cpu_v2)—failure to set this causes deployment errors.

Conclusion: B is correct.

OCI documentation states: “The runtime.yaml file in a model artifact specifies the runtime environment, including the conda environment path (e.g., ENVIRONMENT_SLUG: pyspark30_p37_cpu_v2), ensuring the deployed model uses the correct dependencies.” score.py (A) handles inference, requirements.txt (C) is for pip (not conda in OCI), and D isn’t valid—only B addresses the conda issue per OCI’s deployment process.

Oracle Cloud Infrastructure Data Science Documentation, "Model Deployment - runtime.yaml".

Detailed Answer in Step-by-Step Solution:

Objective: Identify OCI’s managed Kubernetes service.

Evaluate Options:

A: Container Registry—Stores images, not Kubernetes.

B: Load Balancing—Network service, not Kubernetes.

C: Container Engine (OKE)—Managed Kubernetes—correct.

D: Streaming—Data streaming, not containers.

Reasoning: C is OCI’s Kubernetes offering—OKE.

Conclusion: C is correct.

OCI documentation states: “OCI Container Engine for Kubernetes (OKE) (C) provides a managed service to deploy, scale, and manage containerized applications using Kubernetes.” A, B, and D serve other purposes—only C matches per OCI’s container services.

Oracle Cloud Infrastructure OKE Documentation, "Overview".

Detailed Answer in Step-by-Step Solution:

Objective: Identify the required configuration for OCI Logging Analytics to collect logs from an instance.

Understand Logging Analytics: Collects and analyzes logs from OCI resources via Management Agents.

Key Concepts:

Entity: Represents the instance (e.g., Linux VM).

Source: Defines log locations (e.g., file paths).

Log Group: Organizes logs for analysis.

Evaluate Options:

A: Log-Log Group—Groups logs, not collection setup.

B: Entity-Log—Links instance to logs, but not source-specific.

C: Source-Entity—Maps log sources to the instance—correct.

D: Log Group-Source—Post-collection organization, not ingestion.

Reasoning: C establishes the link between the instance and its log sources—key for ingestion.

Conclusion: C is correct.

OCI documentation states: “To collect logs using Logging Analytics, configure a Source-Entity Association (C) to link the Management Agent on the instance (entity) to specific log sources (e.g., file paths).” A and D organize logs post-collection, B is less specific—only C is required for ingestion per OCI’s Logging Analytics setup.

Oracle Cloud Infrastructure Logging Analytics Documentation, "Configuring Log Collection".

Detailed Answer in Step-by-Step Solution:

Step 1: Install, configure, and authenticate Git: Git must be installed and configured (e.g., git config --global user.name) before any repository actions.

Step 2: Configure SSH keys: SSH keys are set up (e.g., ssh-keygen) and added to the remote service (e.g., GitHub, OCI Code Repository) for secure access.

Step 3: Create local and remote Git repository: Initialize a local repo (git init) and create/link a remote repo (e.g., git remote add origin).

Step 4: Commit files: Add files (git add .) and commit them locally (git commit -m "message").

Step 5: Push to remote: Push local commits to the remote repo (git push origin main).

Evaluate Options: Only D (1, 2, 3, 4, 5) follows this logical sequence; others (e.g., A starts with SSH before Git installation) are illogical.

The standard Git workflow in OCI Data Science or general practice begins with installing Git (1), configuring SSH for secure access (2), creating repositories (3), committing locally (4), and pushing remotely (5). The OCI Code Repository documentation aligns with this: “First, install Git and configure authentication (e.g., SSH), then set up repositories and manage code.” D is the only option reflecting this industry-standard process.

Oracle Cloud Infrastructure Code Repository Documentation, "Git Workflow Basics".

Detailed Answer in Step-by-Step Solution:

Objective: Identify the function that measures the difference between predicted and actual values in machine learning.

Understand ML Functions:

Optimizer function: Adjusts model parameters to minimize error (e.g., gradient descent)—it uses the cost, not defines it.

Fit function: Trains the model by fitting it to data—process-oriented, not a measure.

Update function: Typically updates weights during training—not a standard term for error measurement.

Cost function: Quantifies prediction error (e.g., MSE, cross-entropy)—directly represents the difference.

Evaluate Options:

A: Optimizer minimizes the cost, not the cost itself—incorrect.

B: Fit executes training, not error definition—incorrect.

C: Update is vague and not a standard ML term for this—incorrect.

D: Cost function (e.g., loss) measures prediction vs. target—correct.

Reasoning: The cost function (or loss function) is the mathematical representation of error, guiding optimization.

Conclusion: D is the correct answer.

In OCI Data Science, the documentation explains: “The cost function (or loss function) measures the difference between the model’s predicted values and the actual target values, such as mean squared error for regression or cross-entropy for classification.” Optimizers (A) use this to adjust weights, fit (B) is a training step, and update (C) isn’t a defined function here—only the cost function (D) fits the description. This aligns with standard ML terminology and OCI’s AutoML processes.

Oracle Cloud Infrastructure Data Science Documentation, "Machine Learning Concepts - Cost Functions".

Detailed Answer in Step-by-Step Solution:

Objective: Identify IAM components in OCI.

Understand IAM: Manages access via policies, groups, etc.

Evaluate Options:

A: Policies—Core IAM component—correct.

B: Subnets—Networking, not IAM.

C: Instances—Compute, not IAM.

D: VCN—Networking, not IAM.

Reasoning: Only A is an IAM element—others are resources.

Conclusion: A is correct.

OCI documentation states: “IAM includes components like policies (A), groups, and compartments to control resource access.” B, C, and D are infrastructure, not IAM—only A fits per OCI’s IAM framework.

Oracle Cloud Infrastructure IAM Documentation, "IAM Components".

Detailed Answer in Step-by-Step Solution:

Objective: Identify a step unique to MLOps vs. DevOps.

Compare MLOps and DevOps:

DevOps: Focuses on software deployment (CI/CD).

MLOps: Extends DevOps to ML, adding model-specific steps.

Evaluate Options:

A: Continuous deployment—Common to both (software/models).

B: Continuous integration—Common to both (code merging).

C: Continuous delivery—Common to both (releasing updates).

D: Continuous training—Unique to MLOps (retraining models with new data).

Reasoning: Only D addresses ML-specific needs (model retraining).

Conclusion: D is correct.

OCI documentation notes: “MLOps extends DevOps with continuous training, a process unique to machine learning where models are retrained with new data to maintain performance.” CI (B), CD (A), and delivery (C) are shared with DevOps—only continuous training (D) is MLOps-specific.

Oracle Cloud Infrastructure Data Science Documentation, "MLOps Concepts".

Detailed Answer in Step-by-Step Solution:

Objective: Sequence OCI AutoML pipeline stages.

Stages:

Adaptive sampling: Reduces data size if large.

Feature selection: Picks relevant features.

Algorithm selection: Chooses best model type.

Hyperparameter tuning: Optimizes model params.

Evaluate: C (sampling, features, algorithms, tuning) matches logical flow—data first, then model.

Reasoning: Sampling precedes feature work—standard in OCI.

Conclusion: C is correct.

OCI documentation states: “AutoML pipeline runs 1) adaptive sampling, 2) feature selection, 3) algorithm selection, 4) hyperparameter tuning (C).” Sampling reduces data first, then features and models are optimized—other orders (A, B, D) misalign with OCI’s sequence.

Oracle Cloud Infrastructure AutoML Documentation, "Pipeline Sequence".

Detailed Answer in Step-by-Step Solution:

Objective: Spot the invalid Security Zone policy.

Understand Security Zones: Enforce strict security—resources can’t leave easily.

Evaluate Options:

A: Boot volume move—Not allowed; violates zone rules—invalid.

B: Compute no-move—True, enforces security.

C: No public access—True, standard policy.

D: Auto-backup—True, required for compliance.

Reasoning: A contradicts Security Zone immutability.

Conclusion: A is incorrect.

OCI documentation states: “Resources like boot volumes (A) cannot be moved from a Security Zone to a standard compartment—B, C, and D are enforced policies.” A violates OCI’s Security Zone restrictions—only A is not valid.

Oracle Cloud Infrastructure Security Zones Documentation, "Policy Restrictions".

Detailed Answer in Step-by-Step Solution:

Define Git: Git is a version control system—centralized vs. distributed is key.

Evaluate Options:

A: Incorrect—Git is distributed, not centralized (e.g., SVN is centralized).

B: Correct—Distributed, tracks file changes across local and remote repos.

C: Incorrect—Git allows simultaneous contributions; it manages, not prevents, merges.

D: Incorrect—Centralized is wrong, and “copious data” is vague.

Reasoning: Git’s distributed nature (each user has a full repo copy) and change-tracking are core traits.

Conclusion: B is accurate.

OCI documentation aligns with Git’s official definition: “Git is a distributed version control system that tracks changes to files, enabling collaboration and version history management.” A and D misclassify it as centralized, while C misrepresents merge handling—B captures Git’s essence as used in OCI Data Science.

Oracle Cloud Infrastructure Code Repository Documentation, "Git Overview".

Detailed Answer in Step-by-Step Solution:

Understand Model Catalog: The OCI Model Catalog stores trained models and associated details.

Identify Shareable Artifacts: When sharing, all components—model file, metadata (e.g., name, description), hyperparameters (e.g., learning rate), and metrics (e.g., accuracy)—are accessible.

Evaluate Options:

A: Excludes the model itself—incorrect.

B: Excludes metrics—incorrect.

C: Excludes metadata and hyperparameters—incorrect.

D: Includes all components—correct.

Conclusion: D is comprehensive and accurate.

The OCI Model Catalog allows sharing of the model artifact (the trained model), metadata, hyperparameters, and performance metrics, enabling full examination by colleagues. This isdetailed in the official documentation. (Reference: Oracle Cloud Infrastructure Data Science Documentation, "Model Catalog Overview").

Detailed Answer in Step-by-Step Solution:

Define OCI Secrets: Secrets are sensitive data (e.g., API keys, passwords) managed securely in OCI.

Evaluate Options:

A: Object Storage is for general data, not secure secret management.

B: Vault is OCI’s service for storing and managing secrets securely.

C: Autonomous Data Warehouse is for analytics, not secret storage.

D: Oracle Databases store data, not OCI-specific secrets.

Reasoning: Vault is purpose-built for secrets with encryption and access control.

Conclusion: B is correct.

OCI Vault “provides a secure, centralized service to store and manage secrets, such as passwords and keys, with encryption at rest and fine-grained access policies.” Object Storage (A), Autonomous Data Warehouse (C), and Oracle Databases (D) serve other purposes—only Vault (B) is designed for secrets per OCI’s security architecture.

Oracle Cloud Infrastructure Vault Documentation, "Secrets Management".

Detailed Answer in Step-by-Step Solution:

Objective: Find the stage where data imbalance (e.g., skewed classes) is identified.

Understand Stages:

Data Modeling: Training models—assumes data is prepared.

Data Monitoring: Post-deployment tracking—not for initial analysis.

Data Exploration: Analyzing data properties (e.g., distributions)—key for imbalance.

Data Access: Retrieving data—no analysis yet.

Evaluate Options:

A: Modeling uses data, doesn’t detect imbalance—incorrect.

B: Monitoring tracks performance, not initial data issues—incorrect.

C: Exploration (e.g., via pandas) reveals imbalances—correct.

D: Access is just retrieval—incorrect.

Reasoning: Imbalance is assessed during exploration (e.g., class counts).

Conclusion: C is correct.

OCI documentation notes: “Data Exploration involves analyzing the dataset to understand its characteristics, such as identifying class imbalances or missing values, using tools like ADS SDK or Jupyter notebooks.” Modeling (A) and Monitoring (B) occur later, while Access (D) is pre-analysis—only Exploration (C) fits this role.

Oracle Cloud Infrastructure Data Science Documentation, "Data Exploration Stage".

Detailed Answer in Step-by-Step Solution:

Objective: Identify two methods to enhance Object Storage durability.

Understand Durability: Ensures data isn’t lost—focus on redundancy and protection.

Evaluate Options:

A: RAID1—Block volume feature, not Object Storage.

B: Encryption—Secures data, not durability.

C: Versioning—Retains old versions, prevents loss—correct.

D: Limit delete—Prevents accidental deletion—correct.

E: Client encryption—Secures, not durability-focused.

Reasoning: C and D directly protect against data loss—durability-focused.

Conclusion: C and D are correct.

OCI documentation states: “Improve Object Storage durability with Versioning (C) to retain previous object versions and by limiting delete permissions (D) to prevent accidental loss.” A isn’t applicable, B and E focus on security—only C and D enhance durability per OCI’s storage features.

Oracle Cloud Infrastructure Object Storage Documentation, "Data Durability Options".