MLS-C01 Practice Test 18 — 15 Questions

Question 1

Drag and drop the steps to set up cross-validation in a SageMaker training job using the built-in XGBoost algorithm in the correct order.

Accepted Answer

Split the data into training and validation folds, then configure the SageMaker estimator with cross-validation parameters (e.g., num_round, eval_metric, num_cv_folds), then execute the training job (fit), then select the best model based on average CV metrics.. Cross-validation requires data splitting, job configuration with CV parameters, execution, and model selection.

Answer

Configure the SageMaker estimator with cross-validation parameters, then split the data into folds, then execute the training job, then select the best model.

Answer

Split the data into folds, then execute the training job, then configure the estimator, then select the best model.

Answer

Execute the training job, then split the data into folds, then configure the estimator, then select the best model.

Question 2

A data engineer is designing a data lake on Amazon S3 that must support both batch and streaming analytics. The data comes in Parquet format and needs to be queryable by Amazon Athena. Which partitioning strategy will optimize query performance and reduce costs?

Accepted Answer

Partition by date and hour for time-based queries. Partitioning by date and hour is optimal for time-series data in Parquet format queried by Athena because it leverages Hive-style partitioning to prune partitions during query execution, drastically reducing the amount of data scanned. This minimizes Athena's cost (which is based on data scanned) and improves query performance by limiting I/O to only the relevant partitions. Parquet's columnar storage further reduces scan volume when queries select only specific columns, making this combination highly efficient for both batch and streaming ingestion patterns.

Answer

Store data as CSV without partitioning for simplicity

Answer

Partition by device_id for granular access

Answer

Use a single partition for all data to simplify management

Question 3

A data scientist is using Amazon SageMaker to train a deep learning model on a large dataset stored in S3. The training job is taking too long. The data scientist wants to reduce training time without changing the model architecture. Which action should they take?

Accepted Answer

Use Pipe mode for data input. Pipe mode streams training data directly from Amazon S3 into the training container as a Linux FIFO pipe, eliminating the need to download the full dataset to the instance's EBS volume before training begins. This dramatically reduces startup time and improves I/O throughput for large datasets, which is exactly the bottleneck described. Since the model architecture is unchanged, Pipe mode is the correct lever to pull for faster training.

Answer

Use a smaller instance type

Answer

Increase the number of epochs

Answer

Decrease the batch size

Question 4

A data scientist is using Amazon SageMaker to train a custom image classification model using a PyTorch script. The training job runs successfully but the model accuracy is lower than expected. The scientist wants to debug the training process by inspecting gradients and layer outputs. Which SageMaker feature should be used to capture this internal state during training?

Accepted Answer

Use SageMaker Debugger to capture tensors and gradients.. SageMaker Debugger is specifically designed to capture internal model state such as tensors, gradients, and weights during training. It allows you to set rules to monitor for issues like vanishing gradients or overfitting, and to save these tensors for later analysis. This directly addresses the need to inspect gradients and layer outputs to diagnose low accuracy.

Answer

Use SageMaker Experiments to track hyperparameters and metrics.

Answer

Use SageMaker Profiler to profile system bottlenecks.

Answer

Use SageMaker Model Monitor to detect data drift.

Question 5

A data scientist is using Amazon SageMaker Debugger to monitor training jobs. The training loss is decreasing but then suddenly spikes. What is the most likely cause and how should it be addressed?

Accepted Answer

Gradient explosion; apply gradient clipping.. A sudden spike in training loss after a period of decreasing loss is a classic symptom of gradient explosion, where gradients become excessively large during backpropagation, causing the model parameters to update erratically. Amazon SageMaker Debugger can monitor tensors and gradients in real time, and applying gradient clipping (e.g., via `max_grad_norm` in PyTorch or `clip_by_global_norm` in TensorFlow) directly addresses this by capping the gradient norm to prevent destabilizing updates.

Answer

Overfitting; apply regularization.

Answer

Learning rate too low; increase learning rate.

Answer

Vanishing gradients; use ReLU activation.

Question 6

A data scientist is training a binary classifier using logistic regression. The dataset has 100 features and 1 million samples. After training, the model achieves AUC of 0.85 on the test set. The business wants to understand which features contribute most to predictions. Which technique should the data scientist use?

Accepted Answer

Use the coefficients of the logistic regression model as feature importance. Logistic regression is inherently interpretable: each feature has a coefficient that represents the change in log-odds of the positive class per unit change in that feature, holding others constant. With 100 features and 1 million samples, the coefficients are stable and directly usable as feature importance. This is the simplest and most appropriate technique for understanding which features contribute most to predictions.

Answer

Use t-SNE to visualize feature importance

Answer

Use a random forest model and its feature importance attribute

Answer

Use Principal Component Analysis (PCA) to find important components

Question 7

A team is training a large deep learning model on Amazon SageMaker. The training job is taking too long and they want to reduce training time without changing the model architecture. Which action is MOST effective?

Accepted Answer

Use a GPU instance (e.g., p3.2xlarge) for training. Deep learning training is dominated by matrix multiplications that are highly parallelizable, and GPUs are designed with thousands of cores optimized for these operations. Switching from a CPU-based instance to a GPU instance like p3.2xlarge can reduce training time by an order of magnitude without changing the model architecture. This is the most direct and effective action for accelerating deep learning training.

Answer

Switch to a compute-optimized instance like c5.4xlarge

Answer

Increase the batch size and learning rate proportionally

Answer

Use SageMaker Automatic Model Tuning with hyperparameter optimization

Question 8

A data scientist is training a linear regression model and notices that the model performs well on training data but poorly on validation data. Which technique should be applied to reduce overfitting?

Accepted Answer

Apply L2 regularization (Ridge). L2 regularization (Ridge) adds a penalty term proportional to the square of the magnitude of the coefficients to the loss function. This shrinks the weights toward zero, reducing the model's sensitivity to individual features and preventing it from fitting noise in the training data, which directly addresses overfitting.

Answer

Increase the number of epochs

Answer

Add more features

Answer

Remove training examples

Question 9

A machine learning team is using Amazon SageMaker to train a large language model. The training script uses PyTorch and the model requires significant memory. The team wants to use model parallelism across multiple GPUs. Which SageMaker feature should they use?

Accepted Answer

SageMaker model parallelism library. The SageMaker model parallelism library is specifically designed to split a single large model across multiple GPUs, enabling training of models that don't fit in a single GPU's memory. It provides automatic and manual partitioning strategies for PyTorch and TensorFlow, making it the correct choice for model parallelism across multiple GPUs. This library handles the communication and synchronization needed when model layers are distributed across devices.

Answer

SageMaker Distributed Training

Answer

SageMaker Horovod

Answer

SageMaker Debugger

Question 10

A team is using SageMaker to train a model with hyperparameter tuning. The training jobs are taking too long. The team wants to reduce time without sacrificing model quality. Which approach should they take?

Accepted Answer

Enable early stopping in the hyperparameter tuning job.. Enabling early stopping in hyperparameter tuning jobs allows SageMaker to automatically stop training jobs that are not improving on the objective metric, saving time and resources without sacrificing model quality. This is a built-in feature that monitors performance and terminates underperforming trials early.

Answer

Use random search instead of Bayesian optimization.

Answer

Increase the maximum number of training jobs.

Answer

Reduce the maximum runtime per training job.

Question 11

A company wants to build a data lake on Amazon S3. The data lake will store raw data in its original format and also transformed data in Parquet. The data is generated by various sources and must be cataloged for discovery. Which service should the company use to automatically discover, catalog, and make the data searchable?

Accepted Answer

AWS Glue Data Catalog. AWS Glue Data Catalog is a fully managed, persistent metadata store that automatically discovers and catalogs data stored in Amazon S3. It uses AWS Glue crawlers to scan data sources, infer schemas, and create table definitions in a central Hive-compatible metastore. This makes the data searchable and immediately queryable by services like Amazon Athena, Amazon Redshift Spectrum, and Amazon EMR. The Data Catalog is the foundational component for building a data lake on AWS because it provides a unified metadata repository across diverse data sources and formats.

Answer

Amazon S3

Answer

Amazon Athena

Answer

Amazon EMR

Question 12

A media company uses SageMaker to train a recommendation model. The training data is stored in an S3 bucket with versioning enabled. The data pipeline updates the training data daily by overwriting objects with new data. Recently, the model's performance degraded, and the team suspects that the training data was corrupted on a specific day. They want to train the model using the data from a previous version. How can the team retrieve the previous version of the training data?

Accepted Answer

Use the S3 GET Object Version API to download the specific version of each object.. The S3 GET Object Version API allows retrieval of a specific version of an object by specifying the version ID. Since the bucket has versioning enabled, each overwrite creates a new version, and the previous version can be downloaded using its version ID. This is the correct method to retrieve the previous version of the training data.

Answer

Restore the bucket from S3 Glacier, which contains the previous version.

Answer

Use S3 Select to query the previous version of the data.

Answer

Enable S3 replication to a different bucket and use the replicated data.

Question 13

A data engineer is building a data pipeline that aggregates customer transaction data. The engineer notices that some transactions have duplicate entries due to a system error. Which approach should the engineer use to identify and remove duplicates based on a unique transaction ID?

Accepted Answer

Use the drop_duplicates method on the transaction ID column. Using drop_duplicates on the transaction ID column is a straightforward and efficient method to remove duplicate rows based on the unique identifier. Option A is incorrect; sorting and checking consecutive rows is a valid but more complex approach, and not as direct as drop_duplicates. Option B is incorrect because fuzzy matching is designed for approximate matches, not exact duplicates. Option C is incorrect because grouping by all columns and summing would aggregate data, potentially losing information, and does not specifically remove duplicate transaction IDs.

Answer

Sort the data by transaction ID and then check consecutive rows for equality

Answer

Use fuzzy matching to find similar transaction IDs

Answer

Group by all columns and aggregate with sum

Question 14

A training job fails with the error shown. The training script expects a file named 'train.csv' in the 'training' channel. What is the most likely cause?

Accepted Answer

The 'train.csv' file is located inside a subfolder within 's3://my-bucket/data/', and the script expects it directly in the channel path.. The error indicates that the training script cannot find 'train.csv' in the expected location. When SageMaker copies data from an S3 channel path (e.g., 's3://my-bucket/data/') to the training instance, it places the contents of that S3 prefix directly into the channel directory (e.g., '/opt/ml/input/data/training/'). If the CSV file is inside a subfolder (e.g., 's3://my-bucket/data/subfolder/train.csv'), the script will not find it at the top level of the channel path, causing a 'file not found' error.

Answer

The S3 bucket policy denies access to the 'train.csv' file.

Answer

The channel name in the input data configuration does not match the script's expected channel name.

Answer

The training script has a bug that prevents it from reading the file.

Question 15

A data scientist is performing EDA on a dataset stored in Amazon S3 using Amazon Athena. The dataset is partitioned by date, and each partition contains CSV files. The data scientist notices that some queries return zero rows for partitions that should have data. Which THREE steps should the data scientist take to troubleshoot? (Choose 3.)

Accepted Answer

Verify that the CSV files exist in the S3 bucket for the specific partition.. Option A is correct because if the underlying CSV objects are missing or stored under a different S3 prefix than the partition location defined in the table, Athena will scan the partition and return zero rows, so verifying object existence and prefix layout is a fundamental first troubleshooting step. Option B is correct because Athena relies on the AWS Glue Data Catalog for partition metadata; if new date partitions were written to S3 but never registered, queries against those partitions return no data, and MSCK REPAIR TABLE (or ALTER TABLE ADD PARTITION) updates the catalog so the partitions become queryable. Option D is correct because a WHERE clause comparing a string-typed partition or column against a numeric literal (or vice versa) can silently match nothing, so checking column data types in the table definition against the values being filtered is essential. Option C is not appropriate here because converting CSV to Parquet is a performance and cost optimization, not a troubleshooting step for missing rows. Option E is not appropriate because a LIMIT clause only caps the number of returned rows and does not trigger partition discovery or repair catalog metadata.

Answer

Convert the CSV files to Parquet format.

Answer

Re-run the query with a LIMIT clause to force partition discovery.