MLS-C01 Practice Exam 67 — 20 Questions

Question 1

A company's ML model is deployed on a SageMaker endpoint. The model's predictions are used in a customer-facing application that requires low latency. Over time, the model's performance degrades due to data drift. What is the most suitable approach to detect this drift automatically?

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

Accepted Answer

Enable Amazon SageMaker Model Monitor to continuously check for data drift. Amazon SageMaker Model Monitor is purpose-built to automatically detect data drift by continuously comparing incoming inference data against a baseline dataset. It computes statistical metrics (e.g., distribution distances like Kolmogorov-Smirnov or Chi-squared) and raises alerts when drift exceeds configurable thresholds, enabling proactive retraining without manual intervention. This directly addresses the need for automated drift detection in a low-latency customer-facing application.

Answer

Set up a CloudWatch alarm on the endpoint's invocation latency

Answer

Periodically retrain the model using all historical data

Answer

Use Amazon S3 events to trigger a Lambda function that compares distributions

Question 2

Which THREE factors should be considered when selecting the appropriate algorithm for a regression problem? (Choose 3.)

Accepted Answer

The number of features relative to the number of samples. Option A is correct because the ratio of features to samples directly affects algorithm choice: with many features relative to samples, high-variance models like unregularized linear regression or deep neural networks overfit, so regularized methods (Ridge/Lasso/Elastic Net) or dimensionality reduction become preferable, while with abundant samples flexible models are viable. Option B is correct because interpretability requirements constrain the model family: if stakeholders need explainable coefficients, linear regression or decision trees are appropriate, whereas black-box models like gradient boosting or neural networks may be unacceptable despite higher accuracy. Option C is correct because non-linear relationships determine whether a linear model suffices or whether polynomial features, kernel methods (e.g., SVR with RBF kernel), tree ensembles, or neural networks are needed to capture the underlying function. Option D is not a modeling factor—training time of day has no bearing on algorithm suitability, only on scheduling and resource availability. Option E is irrelevant; the color of a laptop has no technical relationship to regression algorithm selection.

Answer

The time of day the training will occur

Answer

The color of the data scientist's laptop

Question 3

A company is deploying a SageMaker model for real-time inference. The endpoint must be highly available and cost-effective. Which TWO actions should the company take? (Select TWO.)

Accepted Answer

Deploy the endpoint with at least two instances in different Availability Zones. Deploying a SageMaker endpoint with at least two instances in different Availability Zones (AZs) ensures high availability by eliminating a single point of failure. If one AZ goes down, traffic is automatically routed to the healthy instance in the other AZ, meeting the requirement for a highly available real-time inference endpoint.

Answer

Use managed spot training for inference

Answer

Use GPU instances for all models even if not required

Answer

Use a single large instance to handle peak load

Question 4

An ML engineer is deploying a model to a SageMaker endpoint for real-time inference. The model requires a custom inference script that preprocesses input data and postprocesses predictions. Which SageMaker feature should be used to implement this custom logic?

Accepted Answer

Create a SageMaker model with a custom inference script that includes pre- and post-processing functions. SageMaker allows you to bring your own container or use a pre-built container with a custom inference script that defines `input_fn`, `predict_fn`, `output_fn`, and `model_fn` functions. These functions handle preprocessing of input data, model prediction, and postprocessing of predictions, enabling custom logic for real-time inference endpoints without requiring separate infrastructure.

Answer

Use SageMaker Ground Truth to transform inference requests

Answer

Use SageMaker Processing jobs to preprocess data before inference

Answer

Use a built-in SageMaker algorithm with the default inference code

Question 5

Refer to the exhibit. A data scientist runs the above CLI command to create a SageMaker training job. The job fails with an error 'Unable to read data from s3://bucket/train/'. What is the MOST likely cause?

Accepted Answer

The IAM role does not have permissions to read from the S3 bucket. The error 'Unable to read data from s3://bucket/train/' indicates that the SageMaker training job cannot access the S3 input data. The most common cause is that the IAM role specified in the command does not have the necessary s3:GetObject permission on the S3 bucket or objects. SageMaker uses the IAM role to assume permissions for reading training data, and without proper S3 read access, the job fails at the data loading stage.

Answer

The training image is not accessible

Answer

The instance type does not support the required memory

Answer

The training job is in a different region than the S3 bucket

Question 6

A data engineer is designing a data pipeline that uses Amazon Kinesis Data Streams to ingest sensor data. The data must be processed in real-time, and the results must be stored in Amazon DynamoDB. Which TWO AWS services can be used together to achieve this? (Choose TWO.)

Accepted Answer

Amazon Kinesis Data Analytics. Amazon Kinesis Data Analytics (option B) is correct because it can consume records directly from a Kinesis Data Stream, run continuous SQL or Apache Flink queries over the streaming data in real time, and then write the transformed results to a downstream sink such as DynamoDB. AWS Lambda (option E) is correct because it can be configured as an event source mapping for a Kinesis Data Stream, invoking the function in real time as records arrive so the function can process each batch and write the results to DynamoDB. Together, Kinesis Data Analytics and Lambda provide the real-time processing and DynamoDB storage required by the pipeline. Amazon Athena (A) is an interactive query service for data in S3, not a stream processor. AWS Glue (C) is a serverless ETL service primarily used for batch and catalog operations, not real-time Kinesis stream processing. Amazon S3 (D) is object storage and does not process streaming data in real time.

Answer

Amazon Athena

Answer

AWS Glue

Answer

Amazon S3

Question 7

A company is deploying a model to an Amazon SageMaker endpoint for real-time inference. The model requires a GPU for low-latency predictions. Which instance type should be chosen?

Accepted Answer

ml.g4dn.xlarge. The ml.g4dn.xlarge instance is correct because it includes an NVIDIA T4 GPU, which is required for low-latency real-time inference with deep learning models. GPU instances accelerate matrix operations and parallel processing, reducing inference latency compared to CPU-only instances.

Answer

ml.c5.xlarge

Answer

ml.r5.2xlarge

Answer

ml.m5.large

Question 8

During EDA, a data scientist finds that a feature 'age' has 30% missing values. The dataset has 100,000 rows. Which imputation strategy is most robust if the data is not missing at random (MNAR) and the missingness is related to the age value itself?

Accepted Answer

Create a separate category indicating missingness and impute with a placeholder. When data are missing not at random (MNAR) and missingness is related to the age value itself, creating a separate category for missingness allows the model to capture the systematic difference. Imputing with a placeholder (e.g., -1) and adding an indicator column preserves the missingness pattern. Option A is wrong because mean imputation ignores the systematic difference and reduces variance. Option B is wrong because median imputation similarly fails to account for the non-random missingness. Option C is wrong because random sampling from observed ages does not capture the relationship between missingness and the value.

Answer

Impute with the mean age

Answer

Impute with a random sample from the observed ages

Answer

Impute with the median age

Question 9

A data scientist is using a decision tree algorithm for a classification task. The tree is very deep and achieves 100% accuracy on the training set but performs poorly on the test set. Which technique should the data scientist use to improve generalization?

Accepted Answer

Prune the decision tree.. A deep decision tree that achieves 100% training accuracy but poor test accuracy is overfitting the training data. Pruning the tree removes branches that have little statistical power, reducing complexity and improving generalization to unseen data.

Answer

Add more features to the dataset.

Answer

Reduce the number of training samples.

Answer

Increase the maximum depth of the tree.

Question 10

Which TWO are appropriate techniques for detecting outliers in a dataset during exploratory data analysis?

Accepted Answer

Z-score method (assuming normal distribution)

Answer

One-hot encoding

Answer

Principal component analysis (PCA)

Answer

t-SNE

Question 11

A company runs a daily batch ETL job using AWS Glue that reads from Amazon RDS (MySQL), transforms the data, and writes to Amazon Redshift. The job takes 6 hours and processes 500 GB of data. Management wants to reduce the runtime. Which action would be MOST effective?

Accepted Answer

Increase the number of DPUs allocated to the Glue job. Increasing the number of DPUs (Data Processing Units) allocated to the AWS Glue job directly increases the parallelism of the Spark-based ETL job, allowing it to process the 500 GB of data faster. Since the bottleneck is the Glue job's compute capacity, adding more DPUs reduces runtime without changing the source or target infrastructure.

Answer

Increase the node size of the Redshift cluster

Answer

Use the Redshift COPY command to load data directly from RDS

Answer

Use Amazon RDS with Provisioned IOPS SSD storage

Question 12

A company wants to track and compare metrics from multiple machine learning experiments. Which Amazon SageMaker feature should be used?

Accepted Answer

SageMaker Experiments. SageMaker Experiments is the correct choice for tracking and comparing metrics from multiple machine learning experiments. SageMaker Model Monitor is used to detect data drift, SageMaker Debugger is used to debug training jobs, and SageMaker Ground Truth is used for data labeling. Thus, only option A is correct.

Answer

SageMaker Ground Truth

Answer

SageMaker Model Monitor

Answer

SageMaker Debugger

Question 13

An engineer sees the error in the exhibit when trying to deploy a model from a model registry in SageMaker. What is the MOST likely cause?

Accepted Answer

The model package version does not exist in the registry. The error in the exhibit indicates that the model package ARN does not exist in the registry. This occurs when the model package version has not been successfully created or registered, meaning it does not exist. Option B is therefore correct: the model package version does not exist. Option A would result in an access denied error, not this ARN-not-found error. Option C is incorrect because an 'Approved' status is actually required for deployment, so it would not cause this error. Option D would produce a different error about an existing endpoint configuration or conflict, not a missing ARN.

Answer

The IAM role lacks permission to access the model registry

Answer

The model package is still in 'Approved' status

Answer

The SageMaker endpoint is already deployed with the same model

Question 14

During training of a deep learning model on a GPU instance in SageMaker, the training job fails with an insufficient memory error. Which step should be taken first to resolve this issue?

Accepted Answer

Reduce the batch size. The most direct cause of an out-of-memory (OOM) error during GPU training is that the combined size of the model parameters, activations, and gradients exceeds the GPU's VRAM. Reducing the batch size immediately decreases the memory footprint of activations stored for backpropagation, which is the largest and most tunable memory consumer. This is the first and simplest step to resolve the error without altering the model architecture or training dynamics.

Answer

Add dropout layers

Answer

Use a smaller learning rate

Answer

Use gradient clipping

Question 15

A company uses Amazon SageMaker to train a linear regression model. After training, the model shows high bias on the training set. Which action is MOST likely to reduce bias?

Accepted Answer

Add more features. High bias indicates that the model is underfitting the training data, meaning it is too simple to capture the underlying patterns. Adding more features increases the model's capacity to learn complex relationships, directly addressing underfitting by reducing bias. In SageMaker, this can be done by engineering additional input columns or using feature transformations before training.

Answer

Collect more training data

Answer

Apply L2 regularization

Answer

Deploy the model to a larger instance

Question 16

Refer to the exhibit. A data engineer runs the AWS CLI command to check an object in an S3 bucket. The bucket is part of a data lake and is configured with versioning enabled. However, the output shows "VersionId": null. What is the most likely reason for this?

Accepted Answer

The object was uploaded before versioning was enabled. When versioning is enabled on a bucket, objects uploaded after that point receive a unique VersionId. Objects that existed before versioning was enabled have a VersionId of null — they are treated as the 'null version.' Therefore, an object showing 'VersionId': null was most likely uploaded before versioning was turned on.

Answer

The object is encrypted using SSE-S3, which hides the version ID

Answer

The command must include the --version-id parameter to display the version ID

Answer

Versioning is not enabled on the bucket

Question 17

A data scientist is analyzing a dataset with 1 million records and 20 features. The target variable is continuous. The scientist wants to identify non-linear relationships between features and the target. Which technique is MOST suitable for this purpose during exploratory data analysis?

Accepted Answer

Calculate mutual information scores between each feature and the target.. Mutual information scores can capture any kind of relationship, including non-linear ones, between features and a continuous target. It measures the mutual dependence between variables, making it suitable for identifying non-linear associations. Pearson correlation only detects linear relationships, and PCA is for dimensionality reduction, not directly for feature-target relationship analysis.

Answer

Visualize the correlation matrix heatmap of all features.

Answer

Apply Principal Component Analysis (PCA) and examine the loadings.

Answer

Compute Pearson correlation coefficients between each feature and the target.

Question 18

A company is using Amazon SageMaker to build a binary classification model for customer churn. The dataset is highly imbalanced (90% no churn, 10% churn). Which technique is MOST effective for handling class imbalance?

Accepted Answer

Use SMOTE to generate synthetic samples for the minority class.. SMOTE (Synthetic Minority Oversampling Technique) is the most effective option because it generates synthetic samples for the minority class by interpolating between existing minority instances, thereby balancing the dataset without discarding valuable majority-class data. This approach directly addresses the class imbalance in a binary classification task on SageMaker, improving model recall for the churn class without the information loss caused by undersampling.

Answer

Use accuracy as the evaluation metric.

Answer

Undersample the majority class.

Answer

Train a random forest model instead of logistic regression.

Question 19

Refer to the exhibit. An ML engineer runs the above CLI command to inspect files in an S3 bucket. The training data consists of 200 CSV files, each 1 GB. The engineer plans to use Amazon SageMaker to train a model using this data. What should the engineer do to optimize training performance?

Accepted Answer

Convert the CSV files to Parquet format and use Pipe input mode.. Converting CSV files to Parquet format and using Pipe input mode significantly improves training performance. Parquet is a columnar storage format that reduces I/O by reading only relevant columns, and it is compressed. Pipe input mode streams data directly from S3 to the training algorithm without downloading to EBS, reducing startup time and disk usage. Option A is incorrect because simply increasing the number of instances does not address the inefficiency of reading CSV files; it may help parallelization but not per-instance throughput. Option B is incorrect because Amazon Athena is a query service, not a data transformation tool for SageMaker; converting to CSV with headers does not improve performance. Option C is incorrect because using File input mode copies all files to the training instance's EBS volume, which is slow for 200 GB of data and does not leverage streaming benefits.

Answer

Increase the number of training instances to process files in parallel.

Answer

Use Amazon Athena to transform the data into CSV format with headers.

Answer

Use the File input mode and copy all files to the training instance's EBS volume.

Question 20

A company wants to analyze historical data stored in Amazon S3 using Amazon Athena. The data is in CSV format and is partitioned by date. Which action will provide the best query performance and cost optimization?

Accepted Answer

Convert the data to Parquet format and use the existing partition structure. Converting CSV to Parquet while preserving the existing date partitions combines the two biggest Athena cost/performance levers: columnar storage (Parquet) and partition pruning. Parquet lets Athena read only the columns needed and benefits from compression and predicate pushdown, while the existing year/month/day partitions let Athena skip irrelevant S3 prefixes. Together they deliver the best query performance and lowest bytes-scanned cost.

Answer

Use AWS Glue to compress the CSV files with gzip

Answer

Create an S3 event notification to trigger a Lambda function that warms up Athena

Answer

Keep CSV format but ensure partitions are in the format year=YYYY/month=MM/day=DD