MLS-C01 Practice Exam 64 — 20 Questions

Question 1

Which THREE actions can help reduce the inference latency of a SageMaker endpoint? (Choose three.)

Accepted Answer

Use a larger instance type with more CPU/GPU. Using a larger instance type with more CPU or GPU resources directly increases the computational capacity available for inference. This reduces the time required to process each prediction request, thereby lowering inference latency. SageMaker endpoints scale horizontally and vertically, and vertical scaling (larger instances) is a straightforward way to improve per-request performance.

Answer

Use a multi-model endpoint to share instances across models

Answer

Increase the number of instances in the endpoint

Question 2

Refer to the exhibit. A data scientist is using AWS Glue ETL jobs to process data from a source database. The job logs show repeated timeout errors. Which EDA step should the scientist perform to diagnose the issue?

Accepted Answer

Check the source database table sizes and row counts over time.. The timeout errors indicate that the Glue ETL job is exceeding its configured timeout. To diagnose the root cause, the data scientist should check the source database table sizes and row counts over time (option B). This helps determine if the data volume has increased, causing longer processing times. Option A (testing network connectivity) addresses network issues but not processing delays. Option C (switching job type) may change performance but does not diagnose the cause. Option D (increasing timeout) is a temporary workaround, not a diagnostic step.

Answer

Test network connectivity from the Glue job to the source database using telnet.

Answer

Switch the Glue ETL job type from Spark to Python shell to reduce overhead.

Answer

Increase the Glue job timeout to 600 seconds and rerun.

Question 3

A financial services company is building a model to detect fraudulent credit card transactions. The dataset contains 1 million transactions, with only 0.1% labeled as fraud. The data scientist trains a logistic regression model on the raw dataset and obtains the following results on a held-out test set: accuracy = 99.8%, precision = 50%, recall = 60%, F1 = 0.545. The business requirement is to maximize recall while keeping precision above 80%. Which course of action should the data scientist take to improve the model?

Accepted Answer

Train the model with class weights inversely proportional to class frequencies. Assigning class weights inversely proportional to class frequencies penalizes misclassifications of the minority class (fraud) more heavily during training. This directly addresses the severe class imbalance (0.1% fraud) by forcing the logistic regression model to learn decision boundaries that improve recall, while the weight ratio can be tuned to maintain precision above 80%. Unlike naive resampling, this approach preserves the original data distribution and avoids information loss.

Answer

Use random undersampling of the majority class to balance the dataset

Answer

Collect more historical transaction data and retrain the model

Answer

Apply L2 regularization with a higher penalty to reduce overfitting

Question 4

You are deploying a PyTorch model to a SageMaker endpoint. The model is large (5 GB) and the endpoint is using an ml.c5.2xlarge instance. Inference latency is higher than required. Which change would most effectively reduce latency?

Accepted Answer

Use a GPU instance type such as ml.p3.2xlarge. The primary bottleneck for a large PyTorch model (5 GB) on a CPU instance (ml.c5.2xlarge) is the lack of GPU acceleration for matrix operations and tensor computations. Switching to a GPU instance like ml.p3.2xlarge (with NVIDIA V100 GPUs) offloads the heavy parallel computation to the GPU, drastically reducing per-inference latency for deep learning models, especially those with large parameter counts.

Answer

Reduce the batch size in the inference code

Answer

Decrease the number of model server workers

Answer

Enable SageMaker Elastic Inference

Question 5

A company uses SageMaker built-in BlazingText algorithm for text classification. The model performance is poor on the validation set. The data consists of short documents (average 50 words). Which hyperparameter tuning strategy is most likely to improve performance?

Accepted Answer

Decrease window size from 5 to 2. BlazingText's default window size of 5 may be too large for short documents (average 50 words), causing the model to learn overly broad context that dilutes local semantic patterns. Decreasing the window size to 2 forces the model to focus on tighter word co-occurrences, which is more effective for short text classification where local n-gram signals are critical.

Answer

Increase bucket size from 0 to 1000000

Answer

Increase vector dimension from 100 to 300

Answer

Increase minCount from 1 to 5

Question 6

A data scientist is using Amazon SageMaker Data Wrangler to perform exploratory data analysis on a dataset. The dataset contains a feature 'age' with values ranging from 0 to 120. The data scientist wants to detect outliers. Which built-in transform in Data Wrangler is most appropriate for this task?

Accepted Answer

Handle Outliers. The 'Handle Outliers' transform in Amazon SageMaker Data Wrangler provides built-in methods such as IQR (Interquartile Range) and z-score to detect and handle outliers in numeric features like 'age'. Option B (Scale and Normalize) is used to rescale features but does not detect outliers. Option C (Handle Missing) deals with missing values, not outliers. Option D (Encode Categorical) is for converting categorical variables to numerical, not for outlier detection.

Answer

Scale and Normalize

Answer

Handle Missing

Answer

Encode Categorical

Question 7

A data engineer is designing an ETL pipeline using AWS Glue to process data from Amazon S3 and load it into Amazon Redshift. The pipeline must handle incremental data loads and ensure data consistency. Which THREE features should the engineer use to achieve this? (Choose THREE.)

Accepted Answer

Pushdown predicates to filter partitions in S3. (pushdown predicates) filters S3 partitions, reducing data scanned and enabling efficient incremental loads. Option D (Redshift transactional tables with automatic commit) ensures data consistency during writes. Option E (Glue job bookmarks) tracks processed data, supporting incremental processing. Option B (Glue data preview) is used for development and does not contribute to incremental loading or consistency. Option C (Glue partition filters) is less efficient than pushdown predicates for filtering partitions.

Answer

Glue data preview to validate transformation logic

Answer

Glue partition filters to limit data scanned

Question 8

A company is using Amazon Kinesis Data Streams to ingest real-time clickstream data. The data is consumed by a Kinesis Data Analytics application that runs SQL queries. The application has been failing intermittently with 'ProvisionedThroughputExceededException' errors. Which action should be taken to resolve this issue?

Accepted Answer

Increase the number of shards in the Kinesis data stream.. The 'ProvisionedThroughputExceededException' error indicates that the Kinesis Data Stream's read or write throughput limits have been exceeded. Increasing the number of shards in the stream directly increases the total provisioned throughput, allowing the Kinesis Data Analytics application to consume data without throttling.

Answer

Disable error logging in the Kinesis Data Analytics application.

Answer

Increase the record size in the Kinesis data stream.

Answer

Switch from Kinesis Data Analytics to Kinesis Data Firehose.

Question 9

A data engineer is using Amazon SageMaker Data Wrangler to perform exploratory data analysis on a large dataset stored in S3. The analysis reveals high cardinality in a categorical feature with over 1 million unique values. What is the best approach to handle this before training a model?

Accepted Answer

Use target encoding based on the mean of the target variable per category.. Target encoding (also known as mean encoding) replaces each category with the mean of the target variable for that category, effectively handling high cardinality without exploding dimensionality or imposing ordinal relationships. Option A is wrong because one-hot encoding on a feature with 1 million unique values would create over 1 million columns, making the dataset sparse and computationally expensive. Option B is wrong because label encoding assigns arbitrary integers, which may introduce unintended ordinal relationships and mislead the model. Option C is wrong because dropping the feature could discard valuable predictive information.

Answer

Apply one-hot encoding.

Answer

Use label encoding to convert categories to integers.

Answer

Drop the high-cardinality feature.

Question 10

A data scientist is training an LSTM model for time series forecasting using Amazon SageMaker. The model is overfitting. Which action is LEAST likely to reduce overfitting?

Accepted Answer

Increase the number of LSTM layers. Increasing the number of LSTM layers (Option B) is least likely to reduce overfitting because it increases model complexity, which can actually worsen overfitting. In contrast, adding dropout layers (Option A), reducing the number of hidden units (Option C), and using early stopping (Option D) are all techniques that help reduce overfitting by imposing regularization or limiting training time.

Answer

Add dropout layers

Answer

Reduce the number of hidden units

Answer

Use early stopping

Question 11

A data scientist is performing exploratory data analysis on a dataset with both numerical and categorical features. The scientist wants to visualize the pairwise relationships between numerical features and also see the distribution of each feature. Which type of plot should the scientist use?

Accepted Answer

Pair plot (scatter matrix) with histograms on the diagonal.. A pair plot (scatter matrix) shows pairwise scatter plots for numerical features and histograms on the diagonal for distribution of each feature. Option B is incorrect because a box plot shows distribution of a single feature, not pairwise relationships. Option C is incorrect because a heatmap of the correlation matrix shows only correlation values, not distributions or actual data points. Option D is incorrect because a correlation matrix with numbers only shows correlation coefficients, not individual feature distributions or pairwise scatter patterns.

Answer

Box plot for each feature.

Answer

Heatmap of the correlation matrix.

Answer

Correlation matrix with numbers.

Question 12

A data scientist is using Amazon SageMaker Data Wrangler to explore a dataset. They notice that a feature has a very high correlation (0.95) with the target variable. What should they do to avoid overfitting?

Accepted Answer

Remove the feature from the dataset. The feature with a 0.95 correlation to the target is likely leaking target information (data leakage), which would cause the model to overfit on training data but fail on new data. Removing the feature (Option D) directly addresses the leakage. Option A (L2 regularization) helps with overfitting from noisy features but does not remove the leaked information. Option B (PCA) reduces dimensionality but the leak would still be present in the principal components. Option C (StandardScaler) only normalizes the feature, not removes it. Therefore, the best action is to remove the feature.

Answer

Use L2 regularization in the model

Answer

Apply PCA to reduce dimensionality

Answer

Standardize the feature using StandardScaler

Question 13

A data scientist is training a multiclass classification model to categorize support tickets into 50 categories. The dataset has 100,000 labeled tickets. The scientist uses a random forest classifier with 100 trees. The model achieves 90% accuracy on the test set, but the F1-score for some rare categories is below 0.1. The scientist wants to improve performance on rare categories without significantly reducing overall accuracy. Which approach should the scientist try?

Accepted Answer

Use class_weight='balanced' or compute custom class weights. (use class_weight='balanced' or compute custom class weights) helps the model focus on rare classes by assigning higher penalties to misclassifications of minority classes. Option B (reduce the number of trees to 50) may reduce model capacity and hurt overall performance. Option C (switch to a one-vs-rest logistic regression model) may not handle rare classes well without class weighting. Option A (increase the maximum depth of trees) could lead to overfitting and may not address class imbalance.

Answer

Increase the maximum depth of trees

Answer

Reduce the number of trees to 50 to prevent overfitting

Answer

Switch to a one-vs-rest logistic regression model

Question 14

A data scientist is using SageMaker to train a model. The training job needs to access data in an S3 bucket in a different AWS account. The data scientist has set up proper S3 bucket policies and IAM roles. Which THREE steps are necessary to allow SageMaker to access the cross-account S3 bucket? (Select THREE.)

Accepted Answer

Configure the S3 bucket policy to grant access to the SageMaker execution role ARN from the training account. The S3 bucket policy in the data account must explicitly grant the SageMaker execution role ARN from the training account the necessary permissions (e.g., s3:GetObject, s3:ListBucket). This is the foundational step for cross-account access, as S3 bucket policies are resource-based policies that can specify principals from other AWS accounts.

Answer

Create a VPC endpoint for S3 in the training account

Answer

Use an AWS KMS key to encrypt the data in transit

Question 15

A team is deploying a model for fraud detection. The dataset is highly imbalanced (99% legitimate, 1% fraudulent). They trained a logistic regression model and achieved 99% accuracy on the test set. However, the model fails to detect most fraud cases. Which metric should the team focus on to evaluate the model?

Accepted Answer

Recall. For imbalanced datasets, accuracy is misleading because it can be high even if the model misses all fraud cases. Recall (true positive rate) measures the proportion of actual fraud cases correctly identified. Option A (Mean squared error) is for regression tasks, not classification. Option B (Precision) measures the proportion of predicted fraud cases that are actually fraud, but it may be high even if recall is low. Option D (Accuracy) is high (99%) but does not reflect the model's poor performance on the minority class. Therefore, recall is the most appropriate metric.

Answer

Mean squared error

Answer

Precision

Answer

Accuracy

Question 16

A data scientist is evaluating a binary classification model. The model's confusion matrix shows: True Positives=80, False Positives=20, True Negatives=900, False Negatives=0. Which THREE metrics can be calculated from this confusion matrix? (Choose three.)

Accepted Answer

Precision. Precision is calculated as TP/(TP+FP) = 80/(80+20) = 0.80. This metric measures the proportion of positive identifications that were actually correct, which is directly derivable from the confusion matrix values.

Answer

AUC-ROC

Answer

Root Mean Squared Error (RMSE)

Question 17

A company's ML pipeline uses AWS Step Functions to orchestrate data preprocessing, training, and evaluation. The training step occasionally fails due to a transient error. What is the most robust way to handle this without manual intervention?

Accepted Answer

Implement a retry policy with exponential backoff on the training step in the state machine. AWS Step Functions natively supports retry policies with exponential backoff, which automatically retries failed tasks after a delay that increases progressively. This handles transient errors (e.g., resource contention, network glitches) without manual intervention, making the pipeline robust and self-healing.

Answer

Configure a CloudWatch alarm to notify the team when the step fails

Answer

Use a parallel state to run multiple training instances simultaneously

Answer

Use a custom Lambda function to catch the error and restart the training step

Question 18

A data engineering team is building a real-time clickstream analytics pipeline on AWS. They need to ingest millions of events per second from mobile apps and websites, process them with low latency, and store the results in Amazon S3 for downstream analysis. Which combination of AWS services should the team use to minimize operational overhead while meeting these requirements?

Accepted Answer

Use Amazon Kinesis Data Streams for ingestion, Amazon Kinesis Data Analytics for real-time processing, and Amazon Kinesis Data Firehose to deliver results to Amazon S3.. Amazon Kinesis Data Streams scales to handle millions of events per second with low latency, Kinesis Data Analytics provides real-time processing without managing infrastructure, and Kinesis Data Firehose delivers processed data to Amazon S3 with automatic buffering and compression, minimizing operational overhead. Option A is wrong because Amazon MQ is a managed message broker for standard protocols (e.g., JMS) and does not offer the high-throughput, real-time streaming capabilities required for clickstream analytics. Option B is wrong because, while Kinesis Data Streams works for ingestion, using Amazon EMR with Spark Streaming adds operational overhead for cluster management and scaling, and is less suited for low-latency, serverless processing compared to Kinesis Data Analytics. Option D is wrong because AWS Glue is a batch ETL service, not designed for real-time ingestion, and Amazon RDS is a relational database that cannot handle the throughput and streaming nature of clickstream data; Glue cannot directly ingest streaming data into RDS in real time.

Answer

Use Amazon MQ to ingest streaming data, AWS Lambda to process each message, and save output to Amazon S3.

Answer

Use Amazon Kinesis Data Streams to ingest data, Amazon EMR to process with Spark Streaming, and save output to Amazon S3.

Answer

Use AWS Glue to ingest data into Amazon RDS, then use AWS Glue ETL jobs to transform and load into Amazon S3.

Question 19

A company is building a binary classifier to predict equipment failure. The dataset has 99% negative (no failure) and 1% positive (failure) examples. The data scientist uses a random forest model with default settings. The model achieves 99% accuracy on the test set but fails to identify any actual failures. Which metric should the data scientist use to evaluate the model?

Accepted Answer

Recall. Recall (sensitivity) measures the proportion of actual positive cases correctly identified. With 99% negative examples, a model can achieve 99% accuracy by simply predicting 'no failure' for all instances, but this yields 0% recall for the failure class. Since the goal is to detect rare failures, recall is the appropriate metric to evaluate the model's ability to find positive cases.

Answer

RMSE

Answer

R-squared

Answer

Precision

Question 20

A data scientist needs to perform exploratory data analysis on a 100 GB CSV file stored in Amazon S3. The data is not sensitive. The scientist wants to use SQL queries to filter and aggregate the data without setting up a server or moving the data. Which service should be used?

Accepted Answer

Amazon Athena. Amazon Athena is the correct choice because it is a serverless, interactive query service that allows you to run standard SQL directly on data stored in Amazon S3 without any infrastructure setup. For a 100 GB CSV file, Athena can handle the query workload efficiently by automatically scaling, and it charges only for the data scanned per query, making it ideal for ad-hoc exploratory analysis without moving or transforming the data.

Answer

AWS Glue

Answer

Amazon EMR

Answer

Amazon Redshift Spectrum