Q1. You are a data scientist trying to load data into your notebook session.

You
understand that Accelerated Data Science (ADS) SDK supports leading various data
formats.
Options:
A) JSON
B) DOCX
C) Raw Images
D) Pandas DataFrame
E) XML
✅ Correct Answer: A) JSON, C) Raw Images, E) XML

Q2. What is the purpose of continuous training in MLOps?

Options:
A) To replace DevOps practices
B) To eliminate the need for data validation
C) To manually update software systems
D) To retrain machine learning models for redeployment
✅ Correct Answer: D) To retrain machine learning models for redeployment

Q3. You want to write a Python script to create a collection of projects for your
data science team. Which interface would you use?
Options:
A) OCI Console
B) Command Line Interface (CLI)
C) OCI Software Development Kit (SDK)
D) Mobile App
✅ Correct Answer: C) OCI Software Development Kit (SDK)

Q4. You want to create a heatmap that displays correlations between continuous and
categorical features. Which ADS SDK method is appropriate?
Options:
A) cramersv_plot()
B) corr()
C) correlation_ratio_plot()
D) pearson_plot()
✅ Correct Answer: C) correlation_ratio_plot()

Q5. You received a new dataset and want to quickly find summary info like number of
features and distributions. Which ADS SDK method do you use?
Options:
A) show_in_notebook()
B) compute()
C) show_corr()
D) to_xgboost()
✅ Correct Answer: A) show_in_notebook()

Q6. You want to install a list of Python packages while creating your Data Science
notebook. Which option allows that?
Options:
A) Using runtime configuration
B) Using storage mounts
C) Invoking public endpoint
✅ Correct Answer: A) Using runtime configuration

Q7. What is the main advantage of using a Resource Principal for authentication in
OCI Data Science?
Options:
A) Prevents the need for policies in IAM
B) Eliminates need to store credentials manually
C) Provides unrestricted access to all resources
D) Only required when using OCI CLI
✅ Correct Answer: B) Eliminates need to store credentials manually

Q8. What is the correct sequence of steps to use OCI Data Science Operator?
Options:
A) Initialize → Install Conda → Configure → Run → Check results
B) Configure → Install Conda → Initialize → Run → Check results
✅ Correct Answer: B) Configure → Install Conda → Initialize → Run → Check results

Q9. What detector in PII Operator is used to obfuscate detected sensitive info?
Options:
A) Mask
B) Anonymize
C) Remove
✅ Correct Answer: A) Mask

Q10. Which Oracle Cloud service provides access to public geospatial images?
Options:
A) Oracle Cloud Infrastructure Data Science
B) Oracle Open Data
C) Oracle Analytics Cloud
D) Oracle Big Data Service
✅ Correct Answer: B) Oracle Open Data

Q11. You are building a model and creating new variables like time of day
(morning/afternoon). Which part of Data Science life cycle is this?
Options:
A) Model validation
B) Model type selection
C) Feature engineering
D) Data access
✅ Correct Answer: C) Feature engineering

Q12. Which THREE annotation classes are supported by OCI Data Labeling?
Options:
A) Key-point and Landmark
B) Polygon Segmentation
C) Classification (Single/Multi-label)
D) Object Detection
E) Named Entity Extraction
F) Semantic Segmentation
✅ Correct Answer: C) Classification (Single/Multi-label), D) Object Detection, E)
Named Entity Extraction

Q13. For labeling legal documents, what annotation classes are available?
Options:
A) Single, Multiple, Entity Extraction
B) Single, Multiple, Object Detection
C) Single, Multiple, Key Value
✅ Correct Answer: A) Single, Multiple, Entity Extraction

Q14. You are creating a model training job expected to take different
hyperparameter values. What’s the most efficient way?
Options:
A) Create new job each time with code changes
B) Create new job and set parameters as environment variables
C) Create code to accept parameters as env vars or CLI args set per run
D) Create new job for each CLI run
✅ Correct Answer: C) Create code to accept parameters as env vars or CLI args set
per run

Q15. You have a complex Python project with subfolders and classes. How to run it
as a Job?
Options:
A) ZIP project folder; Jobs detect main top-level automatically
B) ZIP folder; Jobs identify main file automatically
C) ZIP folder, upload as job artifact, and set JOB_RUN_ENTRYPOINT
D) Rewrite as single executable script
✅ Correct Answer: C) ZIP folder, upload as job artifact, and set JOB_RUN_ENTRYPOINT

Q16. You have received machine learning model training code, without clear
information about the optimal shape to run the training.
Options:
A) Start with a random compute shape and monitor the utilization metrics and time
required to finish the model training. Perform model training optimizations and
performance tests in advance to identify the right compute shape before running the
model training as a job.
B) Start with the strongest compute shape Job's support and monitor the Job Run
metrics and time required to complete the model training. Tune the model so that it
utili as much compute resources as possible, even at an increased cost.
C) Start with a smaller shape and monitor the utilization metrics and time required
to complete the model training. If the compute shape is fully utilized, change to
compute that has more resources and re-run the job. Repeat the process until the
processing time does not improve.
D) Start with a smaller shape and monitor the Job Run metrics and time required to
complete the model training. If the compute shape is not fully utilized, tune the
model parameters, and re-run the job. Repeat the process until the shape resources
are fully utilized.
✅ Correct Answer: C) Start with a smaller shape and monitor the utilization metrics
and time required to complete the model training. If the compute shape is fully
utilized, change to compute that has more resources and re-run the job.

Q17. Autoscaling system making too many adjustments too quickly — what helps
mitigate this?
Options:
A) Custom MQL expressions
B) Cooldown periods
C) Load balancer
D) Static resource allocation
✅ Correct Answer: B) Cooldown periods

Q18. What happens immediately after autoscaling event triggers and cooldown = 10
mins?
Options:
A) Cooldown period begins, no scaling for 10 mins
B) System triggers another scaling event
C) Model reverts to original size
D) Cooldown prevents metrics evaluation
✅ Correct Answer: A) Cooldown period begins, no scaling for 10 mins

Q19. You are creating an OCI Data Science job that trains a model and saves it. How
would you design authentication?
Options:
A) Store personal config file in Vault
B) Package config in job artifact
C) Use pre-authenticated request (PAR)
D) Use Resource Principal of the job run
✅ Correct Answer: D) Use Resource Principal of the job run

Q20. You need to move a Data Science project from Development to Production
compartment. Which statement is correct?
Options:
A) Requires deleting notebooks and models first
B) Cannot move after creation
C) Can move without affecting sessions/models
D) Moving also moves sessions/models
✅ Correct Answer: C) Can move without affecting sessions/models

Q21. You need to make sure that the model you have deployed using AI Quick Actions
is responding with suitable responses.
Options:
A) By evaluating the model
B) By deploying the model
C) By fine-tuning the model
✅ Correct Answer: A) By evaluating the model

Q22. You have built a machine model to predict whether a bank customer is going to
default on a loan. You want to use Local Interpretable Model-Agnostic Explanations
(LIME) to understand a specific prediction.
What is the key idea behind LIME?
Options:
A) Global behavior of a machine learning model may be complex, while the local
behavior may be approximated with a simpler surrogate model.
B) Local explanation techniques are model-agnostic, while global explanation
techniques are not.
C) Model-agnostic techniques are more interpretable than techniques that are
dependent on the types of models.
D) Global and local behaviors of machine learning models are similar.
✅ Correct Answer: A) Global behavior of a machine learning model may be complex,
while the local behavior may be approximated with a simpler surrogate model.

Q23. You are building a machine learning model for predicting loan approvals, and
the client wants to know which features are most influential in determining whether
an applicant is approved.
Which explainability method should you use?
Options:
A) ROC Curve
B) What-If Explanation
C) Individual Conditional Expectation
D) Feature Permutation Importance
✅ Correct Answer: D) Feature Permutation Importance

Q24. You need to build a machine learning workflow that has sequential and parallel
steps. You have decided to use the Oracle Cloud Infrastructure (OCI) Data Science
Pipeline feature.
How is Directed Acyclic Graph (DAG) having sequential and parallel steps built
using Pipeline?
Options:
A) By running a Pipeline
B) Using environmental variables
C) Using Pipeline Designer
D) Using dependencies
✅ Correct Answer: D) Using dependencies

Q25. A data scientist is evaluating a multiclass classification model and notices

that the precision and recall values vary significantly across different class
labels.
What should they use to analyze the model's performance in detail?
Options:
A) Confusion Matrix and Precision-Recall by Label Charts
B) Mean Squared Error (MSE)
C) Residuals QQ Plot
D) ROC Curve
✅ Correct Answer: A) Confusion Matrix and Precision-Recall by Label Charts

Q26. You train a model to predict housing prices for your city. Which two metrics
from the Accelerated Data Science (ADS) ADSEvaluator class can you use to evaluate
the regression model?
Options:
A) Weighted Recall
B) Explained Variance Score
C) Mean Absolute Error
D) Weighted Precision
E) F-1 Score
✅ Correct Answer: B) Explained Variance Score; C) Mean Absolute Error

Q27. You want to ensure that all stdout and stderr from your code are automatically
collected and logged, without implementing additional logging in your code.
How would you achieve this with Data Science Jobs?
Options:
A) On job creation, enable logging and select a log group. Then, select either a
log or the option to enable automatic log creation.
B) Make sure that your code is using the standard logging library and then store
all the logs to Object Storage at the end of the job.
C) You can implement custom logging in your code by using the Data Science Jobs
logging service.
D) Create your own log group and use a third-party logging service to capture job
run details for log collection and storing.
✅ Correct Answer: A) On job creation, enable logging and select a log group. Then,
select either a log or the option to enable automatic log creation.

Q28. You want to build a multistep machine learning workflow by using the Oracle
Cloud Infrastructure (OCI) Data Science Pipeline feature.
How would you configure the conda environment to run a pipeline step?
Options:
A) Configure a block volume.
B) Use command-line variables.
C) Use environmental variables.
D) Configure a compute shape.
✅ Correct Answer: C) Use environmental variables.

Q29. Once LangChain application is deployed to OCI Data Science, what are two ways
to invoke it as endpoint?
Options:
A) Use .invoke() method or Use .predict method
B) Use .predict method or Use CLI
C) Use CLI or Use .invoke()
✅ Correct Answer: C) Use CLI or Use .invoke()

Q30. You are a data scientist leveraging Oracle Cloud Infrastructure (OCI) Data
Science to create a model and need some additional Python libraries for processing
genome sequencing data.
Which of the following THREE statements are correct with respect to installing
additional Python libraries to process the data?
Options:
A) You can install private or custom libraries from your own internal repositories.
B) You can only install libraries using yum and pip as a normal user.
C) You can install any open source package available on a publicly accessible
Python Package Index (PyPI) repository.
D) OCI Data Science allows root privileges in notebook sessions.
E) You cannot install a library that's not preinstalled in the provided image.
✅ Correct Answer: A) You can install private or custom libraries from your own
internal repositories.; C) You can install any open source package available on a
publicly accessible PyPI repository.; (Also: you can install packages as a non-root
user using pip/conda — root access is not allowed.)

Q31. As a data scientist, you are working on a global health data set that has data
from more than 50 countries. You want to encode three features such as 'countries',
'race' and 'body organ' as categories.
Which option would you use to encode the categorical feature?
Options:
A) DataFrameLabelEncoder()
B) OneHotEncoder()
C) show_in_notebook()
D) auto_transform()
✅ Correct Answer: D) auto_transform()

Q32. While reviewing your data, you discover that your data set has a class
imbalance. You are aware that the Accelerated Data Science (ADS) SDK provides
multiple built in automatic transformation tools for data set transformation.
Which would be the right tool to correct any imbalance between the classes?
Options:
A) suggest_recommendations()
B) visualize_transforms()
C) auto_transform()
D) sample()
✅ Correct Answer: C) auto_transform() (Note: sample() can manually rebalance, but
the question asked for automatic transformation tools; auto_transform applies
automatic balancing.)

Q33. During a job run, you receive an error message that no space is left on your
disk device. To solve the problem, you must increase the size of the job storage.
What would be the most efficient way to do this with Data Science Jobs?
Options:
A) Create a new job with increased storage size and then run the job.
B) Edit the job, change the size of the storage of your job, and start a new job
run.
C) Your code is using too much disk space. Refactor the code to identify the
problem.
D) On the job run, set the environment variable that helps increase the size of the
storage.
✅ Correct Answer: B) Edit the job, change the size of the storage of your job, and
start a new job run.

Q34. You are setting up a fine-tuning job for a pre-trained model on Oracle Data
Science. You obtain the pre-trained model from HuggingFace, define the training job
using the ADS Python API, and specify the OCI bucket. The training script includes
downloading the model and dataset.
Which of the following steps will be handled automatically by the ADS during the
job run?
Options:
A) Specifying the replica and shape of instances required for the training job
B) Setting up the conda environment and installing additional dependencies
C) Fetching the source code from GitHub and checking out the specific commit
D) Saving the outputs to OCI Object Storage once the training finishes
✅ Correct Answer: B) Setting up the conda environment and installing additional
dependencies

Q35. You are using a git repository that is stored on GitHub to track your
notebooks. You are working with another data scientist on the same project but in
different notebook sessions.
Which two statements are true?
Options:
A) Only one of you has to clone the GitHub repo as you can share it.
B) You do not have to clone the GitHub repo as you can commit directly from the
notebook session to GitHub,
C) To share your work, you commit it and push it to GitHub. Your coworker can then
pull your changes on to their notebook session,
D) It is a best practice that you and your coworker should work in the same branch
because you are working on the same project.
E) Once you have staged your changes, you run the git commit command to save a
snapshot of the state of your code.
✅ Correct Answer: C) To share your work, you commit it and push it to GitHub. Your
coworker can then pull your changes on to their notebook session.; E) Once you have
staged your changes, you run the git commit command to save a snapshot of the state
of your code.

Q36. As a data scientist, you are working on a movie recommendation application

where you a have a very large movie dataset.
Which Oracle Cloud Infrastructure (OCI) services should you use to develop
interactive Spark applications and deploy Spark workloads?
Options:
A) Data Science and Vault
B) Analytics Cloud and Data Flow
C) Data Flow and Data Science
D) Data Integration and Vault
✅ Correct Answer: C) Data Flow and Data Science

Q37. You are working as a data scientist for a healthcare company. They decide to
analyze the data to find patterns in a large volume of electronic medical records.
You are asked to build a PySpark solution to analyze these records in a JupyterLab
notebook.
What is the order of recommended steps to develop a PySpark application in Oracle
Cloud Infrastructure (OCI) Data Science?
Options:
A) Configure core-site.xml. Install a PySpark conda environment. Create a Data Flow
application with the Accelerated Data Science (ADS) SDK. Develop your PySpark
application. Launch a notebook session.
B) Launch a notebook session. Install a PySpark conda environment. Configure core-
site.xml. Develop your PySpark application. Create a Data Flow application with the
Accelerated Data Science (ADS) SDK.
C) Launch a notebook session. Configure core-site.xml. Install a PySpark conda
environment. Develop your PySpark application. Create a Data Flow application with
the Accelerated Data Science (ADS) SDK.
D) Install a Spark conda environment. Configure core-site.xml. Launch a notebook
session. Create a Data Flow application with the Accelerated Data Science (ADS) SDK
Develop your PySpark application.
✅ Correct Answer: B) Launch a notebook session. Install a PySpark conda
environment. Configure core-site.xml. Develop your PySpark application. Create a
Data Flow application with the Accelerated Data Science (ADS) SDK.

Q38. You realize that your model deployment is about to reach its utilization
limit.
What would you do to avoid the issue before requests start to fail? (Choose TWO)
Options:
A) Update the deployment to add more instances.
B) Update the deployment to use a larger-virtual machine (more CPUs/memory).
C) Delete the deployment.
D) Update the deployment to use fewer instances.
E) Reduce the load balancer bandwidth limit so that fewer requests come in.
✅ Correct Answers: A) Update the deployment to add more instances.; B) Update the
deployment to use a larger-virtual machine (more CPUs/memory).

Q39. When preparing your model artifact to save it to the Oracle Cloud
Infrastructure (OCI) Data Science model catalog, you create a score.py file,
What is the purpose of the score.py file?
Options:
A) Execute the inference logic code.
B) Define the inference server dependencies.
C) Configure the deployment infrastructure.
D) Define the compute scaling strategy.
✅ Correct Answer: A) Execute the inference logic code.

Q40. You are a data scientist working for a manufacturing company. You have
developed a forecasting model to predict the sales demand in the upcoming months.
You created a model artifact that contained custom logic requiring third party
libraries.
When you deployed the model it failed to run because you did not include all the
third party dependencies in the model artifact.
What file should be modified to include the missing libraries?
Options:
A) score.py
B) model_artifact_validate.py
C) requirements.txt
D) runtime.yaml
✅ Correct Answer: D) runtime.yaml

Q41. You are a data scientist working for a utilities company. You have developed
an algorithm that detects anomalies from a utility reader in the grid. The size of
the model artifact is about 2 GB, and you are trying to store it in the model
catalog.
Which three interfaces could you use to save the model artifact into the model
catalog?
Options:
A) Oracle Cloud Infrastructure (OCI) Command Line Interface (CLI)
B) OCI Python SDK
C) ODSC CLI
D) Console
E) Git CLI
F) Accelerated Data Science (ADS) Software Development Kit (SDK)
✅ Correct Answers: A) OCI CLI; B) OCI Python SDK; F) Accelerated Data Science (ADS)
SDK

Q42. After you have created and opened a notebook session, you want to use the
Accelerated Data Science (ADS) SDK to access your data and get started with an
exploratory data analysis.
From which two places can you access or install the ADS SDK?
Options:
A) Oracle Autonomous Data Warehouse
B) Oracle Big Data Service
C) Oracle Machine Learning (OML)
D) Conda environments in Oracle Cloud Infrastructure (OCI) Data Science
E) Python Package Index (PyPi)
✅ Correct Answers: D) Conda environments in OCI Data Science; E) Python Package
Index (PyPi)

Q43. You loaded data into Oracle Cloud Infrastructure (OCI) Data Science. To
transform the data, you want to use the Accelerated Data Science (ADS) SDK. When
you applied the get_recommendations () tool to the ADSDataset object, it showed you
user-detected issues with all the recommended changes to apply to the dataset.
Which option should you use to apply all the recommended transformations at once?
Options:
A) fit transform()
B) get transformed_dataset()
C) auto_transform()
D) visualize_transforms()
✅ Correct Answer: C) auto_transform()

Q44. A data science team is using OCI Vault for storing secrets. They have rotated
a secret's contents, but their application still works without changes. Why?
Options:
A) OCI Vault automatically updates code configurations.
B) The application is caching the previous credentials.
C) The old version of the secret is still in use.
D) The secret's OCID remains the same, allowing automatic updates.
✅ Correct Answer: D) The secret's OCID remains the same, allowing automatic
updates.

Q45. Select two reasons why it is important to rotate encryption keys when using
Oracle Cloud Infrastructure (OCI) Vault to store credentials or other secrets.
Options:
A) Periodically rotating keys make it easier to reuse keys.
B) Key rotation reduces risk if a key is ever compromised.
C) Key rotation improves encryption efficiency.
D) Key rotation allows you to encrypt no more than five keys at a time.
E) Periodically rotating keys limits the amount of data encrypted by one key
version.
✅ Correct Answers: B) Key rotation reduces risk if a key is ever compromised.; E)
Periodically rotating keys limits the amount of data encrypted by one key version.

Q46. You are a data scientist designing an air traffic control model, and you
choose to leverage Oracle AutoML. You understand that the Oracle AutoML. pipeline
consists of multiple stages and automatically operates in a certain sequence.
What is the correct sequence for the Oracle AutoML pipeline?
Options:
A) Adaptive sampling, Feature selection, Algorithm selection, Hyperparameter tuning
B) Algorithm selection, Feature selection, Adaptive sampling, Hyperparameter tuning
C) Algorithm selection, Adaptive sampling, Feature selection, Hyperparameter tuning
D) Adaptive sampling, Algorithm selection, Feature selection, Hyperparameter tuning
✅ Correct Answer: C) Algorithm selection, Adaptive sampling, Feature selection,
Hyperparameter tuning

Q47. You have trained three different models on your data set using Oracle AutoML.
You want to visualize the behavior of each of the models, including the baseline
model, on the test set.
Which class should be used from the Accelerated Data Science (ADS) SDK to visually
compare the models?
Options:
A) ADSExplainer
B) ADSEvaluator
C) ADSTuner
D) EvaluationMetrics
✅ Correct Answer: B) ADSEvaluator

Q48. Which approach does Oracle AutoML use to avoid the cold start problem?
Options:
A) Genetic evolutionary algorithms to evolve new models dynamically
B) Meta-learning to predict algorithm performance on unseen data sets
C) Exhaustive grid search to evaluate every possible model configuration
D) Randomized hyperparameter tuning to generate diverse models
✅ Correct Answer: B) Meta-learning to predict algorithm performance on unseen data
sets

Q49. You have created a conda environment in your notebook session. This is the
first time you are working with published conda environments. You have also created
an Object Storage bucket with permission to manage the bucket.
Which two commands are required to publish the conda environment?
Options:
A) odsc conda publish slug <SLUG>
B) conda activate /home/datascience/conda/<SLUG>/
C) odsc conda create file manifest.yaml
D) odsc conda init-bucket --namespace <NAMESPACE> --bucket-name <BUCKET>
E) odsc conda list --override
✅ Correct Answers: D) odsc conda init-bucket --namespace <NAMESPACE> --bucket-name
<BUCKET>; A) odsc conda publish slug <SLUG>

Q50. Which two statements are true about published conda environments?
Options:
A) They are curated by Oracle Cloud Infrastructure (OCI) Data Science.
B) You can only create a published conda environment by modifying a Data Science
conda environment.
C) In addition to service job run environment variables, conda environment
variables can be used in Data Science Jobs.
D) Your notebook session acts as the source to share published conda environments
with team members.
E) The odsc conda init command is used to configure the location of published conda
environments.
✅ Correct Answers: C) In addition to service job run environment variables, conda
environment variables can be used in Data Science Jobs.; D) Your notebook session
acts as the source to share published conda environments with team members.