MLS-C01 Timed Practice Test 2 — 10 Questions

Question 1

A machine learning engineer is deploying a model for real-time inference using Amazon SageMaker. The model is a large ensemble that requires 8 GB of memory and 4 vCPUs. The expected traffic is 100 requests per second with a 200 ms latency requirement. Which instance configuration should they choose?

10 minute time limit — choose an answer to begin.

Accepted Answer

ml.c5.2xlarge (8 vCPU, 16 GB). (ml.c5.2xlarge) is correct because it offers 8 vCPUs and 16 GB memory, meeting both the compute and memory requirements while being cost-effective for CPU-based inference. Option A (ml.t2.medium) is wrong because it has only 2 vCPUs and 4 GB, insufficient for the 8 GB memory need. Option C (ml.p3.2xlarge) is wrong because it includes a GPU, which is unnecessary and overkill for this CPU-bound workload, and costs more. Option D (ml.m5.large) is wrong because it has only 2 vCPUs and 8 GB memory, lacking the required vCPUs and barely meeting memory.

Answer

ml.t2.medium (2 vCPU, 4 GB)

Answer

ml.p3.2xlarge (8 vCPU, 61 GB GPU)

Answer

ml.m5.large (2 vCPU, 8 GB)

Question 2

A data scientist is exploring a dataset with many features and suspects that some features are highly correlated. Which TWO methods can the scientist use to detect and handle multicollinearity before building a linear regression model?

Accepted Answer

Compute Variance Inflation Factor (VIF) for each feature and remove features with VIF > 10.. Option C is correct because the Variance Inflation Factor quantifies how much the variance of an estimated regression coefficient is inflated due to collinearity with other predictors, and a common rule of thumb is that VIF > 10 (or sometimes > 5) indicates problematic multicollinearity that warrants removing or combining the offending features. Option E is correct because Ridge regression adds an L2 penalty (lambda * sum of squared coefficients) to the ordinary least squares loss, which shrinks correlated coefficients toward zero and stabilizes estimates, making it a standard way to handle multicollinearity without dropping features. Option A is not appropriate as stated because although PCA can decorrelate features, using all components retains the same dimensionality and does not by itself address multicollinearity for interpretable regression; typically one drops components or uses PCR. Option B is a preprocessing step for scale sensitivity and does not detect or resolve correlation among features. Option D, stepwise selection, may incidentally drop some correlated variables but is not a reliable or targeted multicollinearity detection/handling method and can introduce other issues such as overfitting and unstable selection.

Answer

Apply Principal Component Analysis (PCA) and use all components.

Answer

Standardize all features to have zero mean and unit variance.

Answer

Use stepwise feature selection.

Question 3

A data scientist wants to understand the relationship between a categorical feature with 3 levels and a continuous target variable. Which visualization is most appropriate?

Accepted Answer

Box plot grouped by category. A box plot grouped by category (Option C) is the most appropriate visualization because it directly compares the distribution of a continuous target variable across the three levels of a categorical feature. It displays median, quartiles, and potential outliers for each group, making it ideal for understanding central tendency, spread, and skewness in a side-by-side comparison.

Answer

Correlation matrix

Answer

Line chart

Answer

Scatter plot

Question 4

A data scientist is performing exploratory data analysis on a dataset with missing values. The dataset contains a column 'income' with 20% missing values. The income distribution is right-skewed. Which imputation method is most appropriate to preserve the skewness?

Accepted Answer

Impute with the median income. Imputing with the median is most appropriate because the income distribution is right-skewed. The median is robust to outliers and extreme values, so it preserves the central tendency and the skewness of the distribution better than the mean, which is pulled upward by high-income outliers. This keeps the imputed values representative of the typical income in the dataset.

Answer

Impute with the mean income

Answer

Drop rows with missing income

Answer

Impute with the mode income

Question 5

A data scientist is analyzing a dataset with a continuous target variable and suspects that the relationship between a predictor and the target is non-linear. Which THREE techniques can the scientist use to explore and model this non-linearity?

Accepted Answer

Add polynomial features (e.g., x^2, x^3) and check if model performance improves.. Options C, D, and E are correct. Adding polynomial features (e.g., x^2, x^3) allows a linear model to capture non-linear relationships. Decision tree regressors naturally model non-linear interactions between predictors and the target. A scatter plot with a LOESS smooth curve visually reveals non-linear patterns in the data. Option A (logistic regression) is incorrect because it is for binary classification, not for exploring non-linearity with a continuous target. Option B (Pearson correlation) only measures linear relationships, so it is not suitable for detecting non-linearity.

Answer

Apply logistic regression to binarize the target.

Answer

Compute the Pearson correlation coefficient between the predictor and target.

Question 6

A company is building a near-real-time dashboard using data from multiple sources. They need to aggregate millions of events per second with sub-second latency. The architecture must be fully managed and minimize operational overhead. Which service should they use for the aggregation layer?

Accepted Answer

Amazon Kinesis Data Analytics for Apache Flink.. Amazon Kinesis Data Analytics for Apache Flink is the correct choice because it provides a fully managed, serverless runtime for Apache Flink, which is designed for stateful stream processing at scale. It can aggregate millions of events per second with sub-second latency using exactly-once semantics and built-in checkpointing, meeting the near-real-time dashboard requirements without any infrastructure management.

Answer

AWS Lambda functions triggered by Kinesis Data Streams.

Answer

Amazon EMR with Spark Streaming.

Answer

Amazon Redshift with materialized views refreshed frequently.

Question 7

A company is using Amazon SageMaker to train a model and wants to track hyperparameter tuning jobs. Which AWS service is BEST suited to store and query metadata such as tuning job configurations and results?

Accepted Answer

Amazon SageMaker Experiments. Amazon SageMaker Experiments is purpose-built for tracking, organizing, and querying metadata from machine learning training runs, including hyperparameter tuning jobs. It automatically captures configurations, metrics, and results, and provides a Python SDK and SDK API to search and compare trials, making it the best choice for this use case.

Answer

Amazon CloudWatch Logs

Answer

Amazon S3 with Amazon Athena

Answer

Amazon DynamoDB

Question 8

A company is building a data lake on Amazon S3 and wants to use AWS Glue to catalog the data. The data includes CSV, Parquet, and JSON files. The team wants to ensure that the Glue crawler can infer the schema correctly and update the Data Catalog when new partitions are added. Which crawler configuration should be used?

Accepted Answer

Create a single crawler that includes all file extensions and set the 'Update all new and existing partitions' option.. A single AWS Glue crawler can handle multiple file formats (CSV, Parquet, JSON) in a data lake on Amazon S3, and enabling 'Update all new and existing partitions' ensures the Data Catalog is refreshed with both new partitions and any schema changes in existing partitions. This configuration maintains a consistent and up-to-date catalog without manual intervention, which is essential for downstream analytics and machine learning workloads.

Answer

Create separate crawlers for each file format and schedule them at different times.

Answer

Use a crawler that only catalogs Parquet files because they are more efficient.

Answer

Use a crawler with 'Update all new and existing partitions' disabled to avoid schema conflicts.

Question 9

A data engineering team is designing a data pipeline to process streaming data from social media feeds. The data must be deduplicated, enriched with customer information from a relational database, and stored in Amazon S3 in Parquet format. Which AWS services should the team use to build this pipeline? (Select TWO.)

Accepted Answer

AWS Glue. Amazon Kinesis Data Streams (E) is correct because it ingests high-throughput streaming data from social media feeds in real time, allowing custom consumers to deduplicate records before further processing. AWS Glue (A) is correct because it can run Apache Spark ETL jobs that enrich the streamed data with customer information from a relational database via JDBC and write the output to Amazon S3 in Parquet format. Together, Kinesis Data Streams handles real-time ingestion and Glue handles the transformation, enrichment, and Parquet conversion. Amazon Kinesis Data Firehose (B) is not selected because it primarily delivers streaming data to destinations without the flexible deduplication and relational enrichment logic required here. Amazon Athena (C) is a query service for data in S3, not a pipeline component for streaming ingestion or ETL. Amazon SageMaker (D) is a machine learning platform and does not provide the streaming ingestion or ETL capabilities needed for this pipeline.

Answer

Amazon Kinesis Data Firehose

Answer

Amazon Athena

Answer

Amazon SageMaker

Question 10

Which TWO of the following are best practices for training deep learning models on Amazon SageMaker? (Select TWO.)

Accepted Answer

Use Pipe input mode to stream data directly from S3 to the algorithm.. Option B is correct because Pipe input mode streams data directly from Amazon S3 to the training algorithm without downloading the full dataset to the instance's disk, which reduces startup time and disk usage for large datasets. Option D is correct because SageMaker managed spot training uses spare EC2 capacity at up to a 90% discount, and SageMaker automatically manages the spot interruptions and resumes training from checkpoints, making it a cost-optimization best practice. Option A is not a general best practice for training itself; SageMaker Processing is for preprocessing jobs, and data augmentation is typically done there or in the training container, but it is not one of the two targeted best practices here. Option C is wrong because storing training data on attached EBS volumes is not a SageMaker best practice; data should reside in S3 and be accessed via File, FastFile, or Pipe modes. Option E is wrong because disabling checkpointing is not a best practice; checkpointing is essential for spot training and for resuming long jobs, and it does not meaningfully slow training.

Answer

Use SageMaker Processing to perform data augmentation before training.

Answer

Store training data on Amazon EBS volumes attached to the training instance.

Answer

Disable checkpointing to improve training speed.