MLS-C01 Practice Exam 63 — 20 Questions

Question 1

A startup is building a data pipeline that ingests data from multiple sources into an Amazon S3 data lake. The data includes CSV files from legacy systems, JSON from web APIs, and Avro from mobile apps. The data must be transformed into Parquet format and cataloged for querying with Amazon Athena. The pipeline must be serverless and minimize operational overhead. The team has decided to use AWS Glue for ETL and cataloging. However, they are concerned about the cost of running Glue jobs continuously. The data arrives in small batches every 10 minutes. Which approach should the team use to minimize cost while meeting the requirements?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Use scheduled Glue jobs to process the data every hour, consolidating multiple batches. Using scheduled Glue jobs every hour to process accumulated data reduces the number of job runs and associated costs, while still providing near-real-time processing (within the hour). Option A is wrong because Lambda functions have limited execution time and memory, making them unsuitable for large-scale transformations. Option B is wrong because Kinesis Data Firehose can directly deliver streaming data to S3, but it does not handle all source formats natively (e.g., CSV, Avro) and additional transformation may be needed. Option D is wrong because a single daily Glue job introduces too much latency for batch arrivals every 10 minutes.

Answer

Use AWS Lambda functions to transform each file upon arrival and store as Parquet

Answer

Use Amazon Kinesis Data Firehose to stream data directly into S3 and use Glue to catalog it

Answer

Use a single daily Glue job to process all data at once

Question 2

A data scientist needs to evaluate a binary classification model's performance. Which metric is most appropriate when the cost of false positives is very high?

Accepted Answer

Precision. Precision is the most appropriate metric when the cost of false positives is very high because it measures the proportion of positive identifications that were actually correct. In binary classification, precision = TP / (TP + FP), so a high precision means very few false positives occur, directly minimizing the costly error type.

Answer

Accuracy

Answer

F1 score

Answer

Recall

Question 3

A machine learning engineer is training a neural network on Amazon SageMaker using a custom Docker container. The training job fails with an error: 'CUDA out of memory.' The training instance is an ml.p3.2xlarge with 16 GB GPU memory. The model and data fit into memory when using batch size 32, but the engineer wants to maximize GPU utilization. Which approach should the engineer use to fix the out-of-memory error while maintaining efficient training?

Accepted Answer

Implement gradient accumulation with a larger effective batch size. Gradient accumulation allows the engineer to simulate a larger effective batch size by accumulating gradients over multiple forward/backward passes before performing an optimizer step. This keeps the per-step memory footprint low (avoiding CUDA out-of-memory) while maintaining training dynamics similar to a larger batch, thus maximizing GPU utilization without crashing.

Answer

Enable mixed precision training

Answer

Reduce batch size to 1

Answer

Use a CPU-only instance

Question 4

An e-commerce company uses Amazon Kinesis Data Firehose to deliver clickstream data to an Amazon S3 bucket. The data is then queried using Amazon Athena. The marketing team wants to run daily reports that aggregate click events by product ID. However, the reports are slow because Athena scans the entire dataset each time. The data is partitioned by date (e.g., s3://bucket/clickstream/2023/01/01/). The product ID is a column within the data. The data engineering team wants to improve query performance without moving the data to another service. Which approach should the team take?

Accepted Answer

Repartition the data by product ID in addition to date. Repartition the data by product ID in addition to date. This adds a partition level for product ID, so queries that filter on product ID will only scan the relevant partitions. Option A (convert to Parquet) reduces data scanned due to columnar storage and compression, but without partition pruning on product ID, Athena would still scan all partitions for each query. Option B (Redshift Spectrum) would still require scanning data, and involves additional service complexity. Option C (create a view) does not change physical storage; it only provides a logical filter, but Athena still scans all underlying data. Therefore, repartitioning by product ID provides the most direct improvement for queries filtering by product ID.

Answer

Convert the data from JSON to Parquet format

Answer

Use Amazon Redshift Spectrum to query the data

Answer

Create a view in Athena that filters by product ID

Question 5

A data scientist is analyzing a dataset of online retail transactions. The dataset contains 500,000 rows and 10 columns: 'TransactionID', 'CustomerID', 'ProductID', 'Quantity', 'UnitPrice', 'TransactionDate', 'PaymentMethod', 'ShippingAddress', 'Country', and 'TotalAmount'. The data scientist loads the data into a SageMaker notebook and performs initial EDA. The data scientist finds that 'UnitPrice' has a range from $0.01 to $10,000, with a mean of $50 and a median of $20. 'Quantity' ranges from -10 to 100, with negative values indicating returns. 'TotalAmount' is calculated as Quantity * UnitPrice. The data scientist also notices that 2% of the 'CustomerID' values are missing, and 1% of 'ProductID' values are missing. There are no missing values in other columns. The data scientist wants to clean the data and prepare it for customer segmentation. Which course of action is most appropriate?

Accepted Answer

Remove rows with negative 'Quantity' to focus on purchases. Impute missing 'CustomerID' and 'ProductID' with a placeholder such as 'Unknown'.. The most appropriate approach. Negative quantities represent returns, which should be removed when analyzing purchase behavior for customer segmentation. Imputing missing 'CustomerID' and 'ProductID' with a placeholder like 'Unknown' retains data without guessing categorical values. Option A is incorrect because mean imputation is not valid for categorical 'CustomerID'. Option B is incorrect because removing all rows with missing values would discard valuable data. Option C is incorrect because negative quantities are meaningful returns, not errors, and replacing them distorts the data.

Answer

Impute missing 'CustomerID' with the mean of 'CustomerID' and missing 'ProductID' with the mode.

Answer

Remove all rows with any missing values.

Answer

Keep negative 'Quantity' and treat them as errors; replace them with the median of positive quantities.

Question 6

A data scientist is analyzing a dataset of customer reviews for a retail company. The dataset contains text reviews, star ratings (1-5), and customer metadata. The scientist wants to perform sentiment analysis to classify reviews as positive or negative. During EDA, the scientist uses Amazon SageMaker Data Wrangler to visualize the distribution of star ratings and notices that 90% of reviews are 4 or 5 stars, while only 2% are 1 star. The scientist is concerned about class imbalance. Which approach should the scientist take to address the imbalance before modeling?

Accepted Answer

Use the F1-score as the evaluation metric to account for imbalance.. The F1-score balances precision and recall, making it a suitable metric for imbalanced datasets where accuracy would be misleading due to the majority class dominating. Option A (downsampling) is not ideal because it discards potentially useful data from the majority class. Option B (random oversampling) can lead to overfitting by duplicating minority class instances, and is not a guaranteed solution. Option C is incorrect because accuracy is not reliable when classes are imbalanced, as high accuracy can be achieved by simply predicting the majority class.

Answer

Downsample the majority class to create a balanced dataset.

Answer

Use random oversampling of the minority class to balance the dataset.

Answer

Use accuracy as the evaluation metric since it is simple.

Question 7

A data scientist is analyzing a dataset with a large number of missing values in several columns. The dataset is stored in an Amazon S3 bucket and is about 5 TB in size. The scientist wants to understand the pattern of missingness (e.g., is it missing completely at random, missing at random, or not missing at random) before deciding on an imputation strategy. The scientist has access to AWS Glue DataBrew and Amazon SageMaker Studio. Which approach should the scientist take to best understand the missing data patterns?

Accepted Answer

Use AWS Glue DataBrew's data quality and missing data reports. AWS Glue DataBrew provides built-in missing data reports that include visualizations such as heatmaps and bar charts to identify patterns of missingness and help determine whether data is MCAR, MAR, or NMAR. Option A is incorrect because SageMaker Data Wrangler, while useful for data preparation, does not have native missingness pattern analysis. Option C is incorrect because AWS Glue ETL jobs require custom PySpark code and are less efficient for exploratory analysis compared to DataBrew's automated reports. Option D is incorrect because while Amazon Athena can query missing values, it lacks pattern analysis capabilities.

Answer

Use Amazon SageMaker Data Wrangler to create a flow and analyze missingness visually

Answer

Use AWS Glue ETL jobs with PySpark to compute missingness statistics

Answer

Use Amazon Athena to run queries to find missing values per column

Question 8

A data scientist is performing EDA on a dataset containing customer transaction records. The dataset includes columns: 'transaction_id', 'customer_id', 'transaction_amount', 'transaction_date', and 'product_category'. The data scientist wants to check for duplicate transactions and identify any suspicious patterns, such as multiple transactions from the same customer on the same day with the same amount. The dataset has 5 million rows. The data scientist is using a SageMaker Studio notebook with a ml.t3.medium instance. The data is stored in S3. What is the most efficient way to perform this analysis?

Accepted Answer

Use a SageMaker Spark processing job with PySpark to aggregate and detect duplicates.. SageMaker Spark processing jobs distribute the workload across multiple nodes, allowing efficient handling of the 5-million-row dataset within the memory limits of the ml.t3.medium instance. Option B (Athena) is less efficient due to query costs and the need for external setup, and it may not be as flexible for custom duplicate detection logic. Option C (pandas) would likely cause out-of-memory errors on the small instance. Option D (DataBrew) is designed for profiling and basic transformations, not for custom duplicate analysis.

Answer

Use Amazon Athena to run SQL queries to find duplicates.

Answer

Load the entire dataset into a pandas DataFrame and use groupby operations.

Answer

Use AWS Glue DataBrew to create a profile and manually inspect.

Question 9

Which THREE actions should be taken to ensure data security when training a model using Amazon SageMaker with data stored in Amazon S3? (Choose 3.)

Accepted Answer

Use a VPC to isolate the SageMaker training job. Using a VPC to isolate the SageMaker training job ensures that the training instance runs within a private network, preventing direct internet access and allowing traffic to flow only through controlled network interfaces. This reduces the attack surface and helps meet compliance requirements for data security.

Answer

Apply an S3 bucket policy that denies all access except from the SageMaker service

Answer

Attach an EBS volume for storing training data

Question 10

A data scientist is using Amazon SageMaker to train a model with a custom Docker container. The training script reads data from an S3 bucket and writes the model artifact to an S3 bucket. The training job fails with a 'NoSuchKey' error. What is the MOST likely cause?

Accepted Answer

The training data path specified in the input data channel is incorrect.. The 'NoSuchKey' error indicates that the specified key (file path) does not exist within the S3 bucket, which occurs when the training data path provided in the input data channel is incorrect. Option A is wrong because, while a script compatibility issue could cause errors, it would not manifest as a 'NoSuchKey' S3 error; it would likely result in a failure during container execution. Option C is wrong because if the Docker image were not available in Amazon ECR, SageMaker would fail with an 'ImageNotFoundException' or similar ECR-related error, not a 'NoSuchKey' S3 error. Option D is wrong because insufficient S3 permissions (e.g., missing s3:GetObject) would result in an 'AccessDenied' error, not 'NoSuchKey'.

Answer

The training script is not compatible with the Docker image.

Answer

The Docker image is not available in Amazon ECR.

Answer

The SageMaker execution role does not have s3:GetObject permission.

Question 11

A team has trained a deep learning model on Amazon SageMaker using a custom Docker container. They want to deploy the model to a SageMaker endpoint for real-time inference. Which format should the model artifacts be in?

Accepted Answer

A single .tar.gz file containing the model files.. Amazon SageMaker requires model artifacts to be packaged as a single .tar.gz file when using a custom Docker container for real-time inference. This compressed archive must contain the model files (e.g., model.pth, model.h5) and any necessary inference code, as SageMaker extracts the archive to the /opt/ml/model directory during deployment. The .tar.gz format ensures consistent extraction and compatibility with SageMaker's inference pipeline.

Answer

A folder on S3 with the model files.

Answer

No format requirement; any file works.

Answer

A .zip file containing the model files.

Question 12

A machine learning team is using Amazon SageMaker to tune hyperparameters for a neural network. They have defined a hyperparameter tuning job with a random search strategy. The training time per job is very long. Which strategy can reduce the total tuning time?

Accepted Answer

Enable early stopping to terminate poorly performing jobs.. Enabling early stopping allows SageMaker to terminate training jobs that are unlikely to produce better results based on the objective metric, which directly reduces total tuning time by freeing up compute resources for more promising hyperparameter combinations. This is especially effective with random search, where many trials may converge slowly or plateau.

Answer

Use a larger instance type for each training job.

Answer

Switch to Bayesian optimization.

Answer

Increase the number of training jobs.

Question 13

A data scientist is analyzing a dataset with missing values in several features. The dataset is large (10 million rows) and stored in an S3 bucket as CSV files. The scientist wants to use AWS Glue to catalog the data and then use Amazon Athena to query it. However, the missing values are causing errors in downstream machine learning models. Which approach should the scientist take to handle missing values during exploratory data analysis?

Accepted Answer

Use AWS Glue ETL jobs with a custom transformation script that uses the AWS Glue library to drop or impute missing values before writing to a new dataset.. AWS Glue ETL jobs can be used with custom scripts to handle missing values by either dropping rows or imputing values using built-in transforms or custom logic. This is ideal for large-scale datasets stored in S3 as CSV files. Glue integrates with the AWS Glue library for transforming data.

Option A (SageMaker Data Wrangler) is more suitable for interactive data preparation and visualization, but not for automated, large-scale ETL processing of 10 million rows.
Option C (Redshift Spectrum) is primarily a query engine that can query data in S3, but it does not provide built-in data cleaning capabilities for missing values; you would need to use SQL functions like COALESCE, but it's not the best approach for comprehensive ETL.
Option D (Athena) is also a query engine and cannot modify the underlying data; it can impute values in query results, but not write transformed data back to S3 as a cleaned dataset without additional steps.

Answer

Use Amazon SageMaker Data Wrangler to create a data flow that imputes missing values and export the transformed dataset to S3.

Answer

Use Amazon Redshift Spectrum with an external table to query the data and use SQL COALESCE to handle missing values on the fly.

Answer

Use Amazon Athena to run SQL queries that impute missing values and write the results to a new table.

Question 14

A company uses AWS Glue ETL jobs to transform data from Amazon RDS for MySQL to Amazon S3. The transformation includes aggregations and joins. The job runs daily and processes approximately 100 GB of data. Recently, the job started failing with memory errors on the worker nodes. Which approach would MOST effectively resolve the issue without changing the logic?

Accepted Answer

Increase the number of workers in the job configuration. Increasing the number of workers distributes the memory load across more nodes, which directly addresses memory errors in a Spark ETL job without altering the transformation logic. AWS Glue Spark jobs process data in memory across workers, and insufficient total memory causes out-of-memory errors when handling 100 GB of data with aggregations and joins.

Answer

Switch from a Spark ETL job to a Python shell job

Answer

Decrease the number of workers to reduce overhead

Answer

Change the worker type from G.2X to G.1X to increase memory per worker

Question 15

A machine learning engineer is designing an automated ML pipeline for training and deploying models. The pipeline must include data validation, model training, hyperparameter tuning, and model deployment. The engineer wants to use AWS services that integrate well and provide version control. Which THREE services should be combined to achieve this? (Choose THREE.)

Accepted Answer

AWS Step Functions. AWS Step Functions orchestrates the ML pipeline by coordinating data validation, model training, hyperparameter tuning, and deployment as a state machine. It integrates natively with Amazon SageMaker for training and tuning jobs, and with AWS CodePipeline for CI/CD, enabling version control and automated transitions between pipeline stages.

Answer

AWS Glue

Answer

Amazon EMR

Question 16

A company uses SageMaker to train a model that processes sensitive customer data. Due to compliance, the training data must be encrypted at rest and in transit, and the model artifacts must be stored in a secured S3 bucket with encryption. Which combination of actions is REQUIRED?

Accepted Answer

Enable AWS KMS encryption for the SageMaker notebook and training job, and use an S3 bucket with default encryption using AWS KMS. It ensures end-to-end encryption: AWS KMS encryption for the SageMaker notebook and training job encrypts data in transit and at rest within the SageMaker environment, while an S3 bucket with default encryption using AWS KMS encrypts the training data and model artifacts at rest. This combination meets compliance requirements for encryption at rest and in transit, as KMS provides envelope encryption with customer-managed keys, and SageMaker automatically uses TLS for data in transit when KMS is enabled.

Answer

Store data in an S3 bucket with AWS CloudHSM integration

Answer

Use an S3 bucket with SSE-S3 and enable SageMaker Internet-facing mode

Answer

Use an S3 bucket with default encryption (SSE-S3) and enable SSL for all connections

Question 17

A data scientist needs to query a 2 TB dataset stored in Amazon S3 using Amazon Athena. The data is in CSV format and is used for exploratory analysis. Queries are currently slow and expensive. Which action will improve query performance and reduce cost?

Accepted Answer

Convert the data to Parquet format and partition by a commonly filtered column.. Converting CSV data to Parquet (a columnar storage format) significantly reduces the amount of data scanned by Athena, as only the columns needed for the query are read. Partitioning by a commonly filtered column (e.g., date or region) further limits the data scanned to only relevant partitions, directly reducing both query cost (Athena charges per TB scanned) and query execution time.

Answer

Convert the data to JSON format to improve compression.

Answer

Increase the number of workers in the Athena query engine.

Answer

Create a composite index on the data using Athena's index feature.

Question 18

Refer to the exhibit. A data scientist is using an IAM role with this policy to run a SageMaker processing job that reads data from S3. The job fails with an access error. What is the most likely cause?

Accepted Answer

The policy does not allow s3:ListBucket. The processing job needs both s3:GetObject and s3:ListBucket to read objects. The policy lacks s3:ListBucket. Option A is wrong because sagemaker:CreateProcessingJob is allowed. Option B is wrong because s3:PutObject is not needed for reading. Option D is wrong because the policy allows s3:GetObject.

Answer

The policy does not allow sagemaker:CreateProcessingJob

Answer

The policy does not allow s3:PutObject

Answer

The policy does not allow s3:GetObject

Question 19

A data scientist ran an XGBoost training job on Amazon SageMaker using a CSV dataset. The training job failed with the error shown. What is the most likely cause of this failure?

Accepted Answer

One of the rows in the CSV has an extra column. XGBoost on SageMaker expects the CSV input to have a consistent number of columns across all rows. If one row contains an extra column, the parser will fail because it cannot map the additional field to the feature schema defined by the training job. This mismatch causes the 'Error: Number of columns does not match' or similar parsing error.

Answer

The dataset includes a header row

Answer

The delimiter used in the CSV is not a comma

Answer

The dataset contains missing values

Question 20

A company is building a binary classification model to predict customer churn. The dataset is highly imbalanced (95% non-churn, 5% churn). The data scientist uses SMOTE to oversample the minority class. After training a logistic regression model, the recall for the churn class is 0.80, but the precision is only 0.10. Which action would MOST likely improve precision without significantly harming recall?

Accepted Answer

Increase the classification threshold for the positive class. Increasing the classification threshold for the positive class makes the model more conservative in predicting churn, which reduces false positives and improves precision. While this may slightly decrease recall (true positives), the trade-off is acceptable given the very low precision. Option A is wrong because random oversampling can lead to overfitting and does not directly address the low precision caused by threshold issues. Option B is wrong because reducing features could discard important information, potentially harming recall without necessarily improving precision. Option D is wrong because decreasing the threshold would further increase false positives, lowering precision even more.

Answer

Use random oversampling instead of SMOTE

Answer

Reduce the number of features in the model

Answer

Decrease the classification threshold for the positive class