MLS-C01 Practice Test 28 — 15 Questions

Question 1

A machine learning team is analyzing a dataset with 10,000 rows and 200 features. They suspect data leakage due to time-based features. Which THREE EDA checks should they perform?

Accepted Answer

Plot distribution of each feature in training vs. test sets. Option A is correct because plotting each feature's distribution in the training versus test sets reveals covariate shift or leakage-driven discrepancies, such as a feature whose test distribution is unrealistically similar to the training target or whose range differs sharply, which is a classic symptom of time-based leakage when random splits are used. Option C is correct because time-based leakage often stems from the data being sorted chronologically or from features that encode future information (e.g., rolling aggregates computed over later timestamps); checking sort order and whether any feature references future events directly detects this. Option D is correct because comparing each feature's correlation with the target in the training versus test sets exposes features that are suspiciously predictive in one split but not the other, a strong indicator that the feature leaks target or future information across the split boundary. Option B is not appropriate because PCA component separation of train/test is a dimensionality-reduction visualization, not a leakage diagnostic, and it would not pinpoint time-based leakage. Option E is not appropriate because k-means clustering on the whole dataset ignores the train/test split and cannot reveal leakage; it only describes cluster structure.

Answer

Apply PCA and check if first two components separate train/test

Answer

Perform k-means clustering on the whole dataset

Question 2

During EDA of a dataset for a regression problem, a data scientist notices that the target variable has a right-skewed distribution. Which THREE transformations are appropriate to address this skewness? (Choose THREE.)

Accepted Answer

Log transformation. Options A, C, and D are correct. Log transformation, Box-Cox transformation, and Yeo-Johnson transformation are effective methods for reducing right skewness in the target variable. Option B (StandardScaler) standardizes features to have zero mean and unit variance, but does not reduce skewness. Option E (Min-Max scaling) scales features to a fixed range, but does not affect the shape of the distribution.

Answer

StandardScaler (z-score normalization)

Answer

Min-Max scaling

Question 3

A data engineer is performing exploratory data analysis on a large dataset stored in Amazon S3 (10 TB in CSV format). The dataset has 2000 columns and 50 million rows. The engineer needs to compute summary statistics (mean, median, standard deviation) for each numeric column and identify missing values. Which approach is MOST cost-effective and time-efficient?

Accepted Answer

Convert the data to Apache Parquet format, then use Amazon Athena to run SQL queries for statistics.. Converting the CSV data to columnar Parquet and querying with Athena is the most cost-effective and time-efficient approach. Parquet is columnar and compressed, so Athena scans only the columns needed for statistics, drastically reducing data scanned and cost. Athena is serverless, so there is no cluster to provision, and it can compute aggregates directly over S3.

Answer

Use Amazon Redshift Spectrum to query the data directly from S3.

Answer

Load the data into Amazon SageMaker Data Wrangler and compute statistics interactively.

Answer

Use AWS Glue ETL to compute statistics and write results to S3.

Question 4

A machine learning engineer is deploying a model to SageMaker for real-time inference. The model is a TensorFlow SavedModel. Which SageMaker capability should be used to create an endpoint?

Accepted Answer

SageMaker hosting with TensorFlow Serving container. SageMaker hosting with TensorFlow Serving container is the correct choice because it provides a managed environment for deploying TensorFlow SavedModels for real-time inference. SageMaker supports custom containers and pre-built containers for popular frameworks like TensorFlow.

Answer

SageMaker Pipelines

Answer

SageMaker Model Monitor

Answer

SageMaker Ground Truth

Question 5

An ML team is using SageMaker Processing jobs to run feature engineering scripts. The scripts require a specific Python package not included in the default SageMaker image. How should the team provide this package?

Accepted Answer

Create a custom Docker image that includes the package and use it for the Processing job. SageMaker Processing jobs run in isolated Docker containers, and the default SageMaker images only include pre-installed packages. To add a custom Python package, the team must create a custom Docker image that includes the package (e.g., via a Dockerfile with 'pip install <package>'), then specify that image URI in the Processing job configuration. This ensures the package is available in the container environment before the script executes.

Answer

Include 'pip install <package>' in the processing script

Answer

Use the SageMaker prebuilt deep learning container with the package

Answer

Place a requirements.txt file in the input data S3 bucket

Question 6

A data scientist is building a model to predict housing prices using a dataset with 100,000 records and 50 features. The features include 'sqft_living', 'sqft_lot', 'bedrooms', 'bathrooms', 'floors', 'waterfront', 'view', 'condition', 'grade', etc. The data scientist uses Amazon SageMaker Data Wrangler for EDA. Upon reviewing the data, the data scientist finds that 'sqft_living' has a correlation of 0.7 with 'sqft_above' (square footage above ground) and 0.6 with 'sqft_basement'. Also, 'grade' (overall grade of the house) is highly correlated with 'condition' (0.8). The target variable 'price' is right-skewed. The data scientist plans to use a linear regression model. Which set of actions should the data scientist take to improve model performance?

Accepted Answer

Apply log transformation to the target variable 'price' to reduce skewness, and remove either 'sqft_above' or 'sqft_living' and either 'grade' or 'condition' to handle multicollinearity.. The target 'price' is right-skewed, so a log transform stabilizes variance and makes the relationship more linear for linear regression. The dataset also has strong multicollinearity (sqft_living vs sqft_above/basement; grade vs condition), which inflates coefficient variance and destabilizes OLS estimates, so dropping one feature from each highly correlated pair is the correct fix. Option D addresses both issues directly.

Answer

Apply standard scaling to all numeric features and use the data as is, since linear regression is robust to multicollinearity.

Answer

Remove all features that have correlation >0.5 with any other feature to eliminate multicollinearity, and apply standard scaling to all numeric features.

Answer

Apply principal component analysis (PCA) to all features to reduce dimensionality, and then fit linear regression on the principal components.

Question 7

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. Amazon SageMaker (E) is the core service for building, training, tuning, and deploying ML models, providing built-in hyperparameter tuning jobs, model registry versioning, and endpoints for deployment. AWS Step Functions (B) orchestrates the multi-step ML workflow (data validation, training, tuning, deployment) as a state machine, coordinating SageMaker jobs and handling retries and branching logic. AWS CodePipeline (C) provides CI/CD automation and version control integration, triggering and managing the pipeline stages from source commits through deployment. AWS Glue (A) is a serverless ETL service for data cataloging and transformation, not a full ML training/tuning/deployment platform, and Amazon EMR (D) is a big-data processing framework for Hadoop/Spark clusters, which is unnecessary here since SageMaker handles the ML lifecycle.

Answer

AWS Glue

Answer

Amazon EMR

Question 8

A data scientist is working with a dataset that includes a 'timestamp' column. They want to create features that capture seasonality. Which feature engineering approach is most appropriate?

Accepted Answer

Extract hour of day and apply sine/cosine transformation.. Extracting hour of day and applying a sine/cosine transformation captures the cyclical nature of time, ensuring that hour 23 and hour 0 are close in the feature space, which is essential for models to learn daily seasonality. This approach avoids the discontinuity of raw integer hours and provides a smooth, periodic representation.

Answer

Bin timestamps into fixed intervals.

Answer

Convert timestamp to Unix epoch seconds.

Answer

One-hot encode the timestamp column.

Question 9

A data scientist is training a deep learning model on a large dataset using SageMaker. The training job is taking too long. Upon reviewing the CloudWatch logs, the scientist notices that the GPU utilization is below 10% most of the time. Which change is MOST likely to improve GPU utilization and reduce training time?

Accepted Answer

Increase the batch size in the training script.. Low GPU utilization (below 10%) indicates that the GPU is idle most of the time, waiting for data to be fed. Increasing the batch size allows each training step to process more samples per forward/backward pass, keeping the GPU busy with larger matrix operations and reducing the relative overhead of data loading and kernel launches. This directly improves GPU throughput and reduces total training time.

Answer

Use a different optimizer that requires less computation.

Answer

Switch to a smaller instance type to reduce data transfer overhead.

Answer

Reduce the size of the training dataset.

Question 10

An ML engineer is troubleshooting why an automated CI/CD pipeline cannot deploy an updated model to an existing SageMaker endpoint. The pipeline uses the IAM role that has the attached policy shown in the exhibit. What is the MOST likely cause of the failure?

Accepted Answer

The pipeline tries to update an existing endpoint, but the sagemaker:UpdateEndpoint action is not allowed.. The pipeline is attempting to deploy an updated model to an existing SageMaker endpoint, which requires the sagemaker:UpdateEndpoint action. The IAM policy shown in the exhibit (not provided here but implied) does not include this action, so the API call fails with an access denied error. Without explicit permission to update the endpoint, the CI/CD pipeline cannot modify the deployed configuration.

Answer

The pipeline tries to create a new endpoint, but the sagemaker:CreateEndpoint action is denied.

Answer

The pipeline tries to delete the old endpoint, but the sagemaker:DeleteEndpoint action is denied by a Deny statement.

Answer

The pipeline attempts to invoke the endpoint, but the sagemaker:InvokeEndpoint action is denied.

Question 11

A team wants to build a data pipeline that processes incoming JSON files from an S3 bucket and loads them into a Redshift table. The pipeline must handle schema evolution and data validation. Which combination of services would be MOST appropriate?

Accepted Answer

Amazon S3 + AWS Glue + Amazon Redshift. AWS Glue provides built-in schema discovery and evolution capabilities via its crawlers and the Data Catalog, which automatically detect and adapt to changes in JSON schemas. Combined with Glue ETL jobs for data validation and transformation, it seamlessly loads processed data into Amazon Redshift, making it the most appropriate choice for handling schema evolution and validation in this pipeline.

Answer

Amazon S3 + Amazon SQS + Amazon Redshift

Answer

Amazon S3 + AWS Data Pipeline + Amazon Redshift

Answer

Amazon S3 + AWS Lambda + Amazon Redshift

Question 12

A data scientist is training a linear regression model and wants to check for multicollinearity among the features. Which TWO methods can be used to detect multicollinearity? (Choose TWO.)

Accepted Answer

Compute the correlation matrix between features. Option B is correct because computing the correlation matrix between features reveals pairwise linear relationships; high absolute correlation coefficients (typically above 0.7 or 0.8) between predictor variables indicate potential multicollinearity. Option D is correct because the Variance Inflation Factor (VIF) quantifies how much the variance of an estimated regression coefficient is inflated due to collinearity with other features, with VIF values above 5 or 10 commonly signaling problematic multicollinearity. Option A is not specific to multicollinearity, since the model's R-squared reflects overall goodness of fit rather than relationships among predictors. Option C is incorrect because coefficient p-values test whether individual coefficients differ from zero, and they can be affected by multicollinearity but do not directly detect it. Option E is incorrect because a residuals vs. fitted values plot is used to assess linearity and homoscedasticity, not multicollinearity among features.

Answer

Examine the R-squared value of the model

Answer

Check the p-values of the coefficients

Answer

Plot the residuals vs. fitted values

Question 13

An organization stores streaming data in Amazon Kinesis Data Streams. A data analyst wants to perform real-time exploratory data analysis on the incoming data to detect anomalies. Which AWS service should the analyst use to run SQL queries on the streaming data?

Accepted Answer

Amazon Kinesis Data Analytics. Amazon Kinesis Data Analytics (now Amazon Managed Service for Apache Flink) is purpose-built to run continuous SQL queries over streaming data in Kinesis Data Streams, enabling real-time anomaly detection without provisioning servers. It ingests the stream, applies SQL transformations or windowed aggregations, and emits results to downstream destinations.

Answer

Amazon SageMaker

Answer

AWS Glue

Answer

Amazon Athena

Question 14

A data scientist is training a neural network on image data using TensorFlow with GPU instances on SageMaker. The training is slow because the GPU utilization is low. The data pipeline uses tf.data with a large number of preprocessing operations. Which action would most likely increase GPU utilization?

Accepted Answer

Increase the prefetch buffer size in the tf.data pipeline.. Low GPU utilization with a heavy tf.data preprocessing pipeline indicates the GPU is starving for data — the input pipeline cannot feed batches fast enough. Increasing the prefetch buffer size allows the CPU-side pipeline to prepare more batches ahead of time and overlap preprocessing with GPU compute, keeping the accelerator busy. This is the standard TensorFlow performance fix for input-bound training.

Answer

Increase the learning rate to converge faster.

Answer

Reduce the batch size to speed up each step.

Answer

Increase the number of CPU instances in the training job.

Answer

Use smaller image sizes to reduce computation.

Question 15

A data scientist is using SageMaker to train a model with a custom algorithm. The training script uses TensorFlow and runs on GPU instances. The training job fails with 'CUDA_ERROR_OUT_OF_MEMORY'. What is the most likely cause?

Accepted Answer

The batch size is too large for the GPU memory. The error 'CUDA_ERROR_OUT_OF_MEMORY' indicates that the GPU memory has been exhausted during training. In TensorFlow, the batch size directly determines how many samples are processed simultaneously on the GPU; a batch size that is too large will exceed the available GPU memory, causing this specific CUDA error. Reducing the batch size is the standard fix for this issue.

Answer

The S3 bucket is in a different region

Answer

The GPU driver is outdated

Answer

The training script has a memory leak on CPU

Answer

The instance type does not have enough CPU cores