MLS-C01 Machine Learning Implementation and Operations Practice Test 5 — 15 Questions

Question 1

A data scientist is deploying a real-time inference endpoint using SageMaker. The model is a large NLP model requiring GPU for low latency. The endpoint must be highly available across two Availability Zones. Which deployment configuration meets these requirements?

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

Accepted Answer

Deploy a multi-model endpoint on an ml.p3.2xlarge instance with auto-scaling and at least two instances in different AZs. It uses a GPU instance (ml.p3.2xlarge) to meet the low-latency requirement for a large NLP model, and it deploys at least two instances across different Availability Zones (AZs) to achieve high availability. SageMaker multi-model endpoints allow hosting multiple models on the same endpoint, but here the key is the instance type and the multi-instance, multi-AZ deployment for fault tolerance.

Answer

Deploy a single model endpoint on an ml.c5.xlarge instance with auto-scaling

Answer

Use SageMaker batch transform on GPU instances

Answer

Deploy a single model endpoint on an ml.p3.2xlarge instance with one instance

Question 2

A machine learning team is using SageMaker Processing jobs to run feature engineering on large datasets. The job takes a long time to complete. Which change would most likely reduce the processing time?

Accepted Answer

Increase the number of instances in the processing cluster. Increasing the number of instances in the processing cluster enables SageMaker Processing to distribute the workload across multiple nodes, leveraging parallel processing to reduce the overall execution time. SageMaker Processing uses a distributed computing model where each instance processes a subset of the data, so adding more instances directly increases parallelism and throughput for embarrassingly parallel tasks like feature engineering.

Answer

Switch to local mode to avoid network overhead

Answer

Change the processing script from Python to PySpark

Answer

Use a larger instance type, e.g., from r5.xlarge to r5.24xlarge

Question 3

A company is using SageMaker to train a model. The training data includes personally identifiable information (PII). The company must ensure that the data is encrypted at rest and in transit. Which combination of actions meets these requirements?

Accepted Answer

Use S3 server-side encryption and enable SageMaker inter-container traffic encryption. It combines S3 server-side encryption (SSE-S3) to protect data at rest in the training dataset with SageMaker inter-container traffic encryption, which encrypts data in transit between training containers using TLS. This dual approach satisfies the requirement for encryption both at rest and in transit within the SageMaker training environment.

Answer

Use S3 server-side encryption with S3 managed keys (SSE-S3)

Answer

Enable SSL for data in transit and use VPC endpoints

Answer

Place all resources in a private VPC subnets with no internet access

Question 4

A data scientist wants to use AWS Step Functions to orchestrate a machine learning workflow including data preprocessing, training, and evaluation. Which SageMaker integration is best suited for this purpose?

Accepted Answer

Use the SageMaker SDK with Step Functions service integrations. AWS Step Functions has a native service integration with SageMaker, allowing you to directly call SageMaker API actions (e.g., CreateTrainingJob, CreateModel, CreateEndpointConfig, CreateEndpoint) without needing intermediate Lambda functions. This integration supports both synchronous and asynchronous invocations, making it ideal for orchestrating a multi-step ML workflow with built-in error handling, retries, and state management.

Answer

Implement each step as an AWS Lambda function and call Step Functions

Answer

Use SageMaker Pipelines to define the workflow

Answer

Use AWS Batch to run the steps sequentially

Question 5

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. Option A is correct because the S3 bucket policy in the data (bucket-owning) account must explicitly grant access to the SageMaker execution role ARN from the training account; cross-account access requires a resource-based policy that names the trusted principal, otherwise S3 will deny the request even if the caller's identity policy allows it. Option C is correct because an IAM role must exist in the data account that defines the permissions to read from the S3 bucket, serving as the cross-account role that the training account's principal will assume. Option E is correct because the SageMaker execution role in the training account must be configured with sts:AssumeRole permission and actually assume the IAM role in the data account, which is the mechanism that establishes the cross-account trust and grants temporary credentials for S3 access. Option B is not required because a VPC endpoint for S3 is only needed for private network access from a VPC and is unrelated to cross-account authorization. Option D is not required because KMS encryption for data in transit is not a prerequisite for cross-account S3 access; TLS handles in-transit encryption, and KMS keys are for at-rest encryption and would only matter if the objects were KMS-encrypted and the role needed decrypt permissions.

Answer

Create a VPC endpoint for S3 in the training account

Answer

Use an AWS KMS key to encrypt the data in transit

Question 6

A data scientist attempts to create a SageMaker training job using the IAM policy shown in the exhibit. The training job fails with an access denied error. What is the most likely cause?

Accepted Answer

The IAM policy is missing the s3:ListBucket action. The IAM policy shown in the exhibit grants s3:GetObject and s3:PutObject actions but is missing the s3:ListBucket action. When SageMaker attempts to read or write objects in an S3 bucket, it first needs to list the bucket's contents to verify the object's existence and path. Without s3:ListBucket, the training job fails with an access denied error during the initial S3 operation.

Answer

The S3 bucket policy does not grant access to the SageMaker service principal

Answer

The IAM policy is missing the sagemaker:DescribeTrainingJob permission

Answer

The IAM policy is missing the s3:PutObject permission

Question 7

A SageMaker training job fails with the failure reason shown in the exhibit. What is the most likely cause?

Accepted Answer

There is an error in the custom training script. The failure reason in the exhibit (not shown here but implied by the question) typically indicates a runtime error such as a Python exception, missing module, or syntax error in the custom training script. SageMaker logs the exact error from the container, and when the script itself fails, the training job terminates with a 'ClientError' or 'AlgorithmError' referencing the script, not infrastructure issues.

Answer

The training instance ran out of memory

Answer

The S3 bucket with training data is not accessible

Answer

The SageMaker service limit for the instance type has been exceeded

Question 8

A data scientist is training a model on Amazon SageMaker and notices that the training job is taking much longer than expected. The instance type is ml.m5.xlarge and the dataset is 10 GB in CSV format. Which action is MOST likely to reduce training time without changing the instance type?

Accepted Answer

Convert the dataset to RecordIO or Parquet format before training.. Converting the dataset from CSV to a columnar format like Parquet or RecordIO reduces I/O overhead and improves data throughput during training. SageMaker's built-in algorithms and many deep learning frameworks can read these formats more efficiently, especially for large datasets, because they enable better compression and predicate pushdown, reducing the time spent on data loading.

Answer

Change the instance type to ml.p3.2xlarge (GPU) for faster computation.

Answer

Reduce the number of training epochs to speed up convergence.

Answer

Increase the batch size to the maximum supported by the instance memory.

Question 9

A machine learning engineer is deploying a real-time inference endpoint using Amazon SageMaker. The model is a large deep learning model that requires low latency (under 100 ms) and high throughput (1000 requests per second). Which SageMaker deployment option is MOST suitable?

Accepted Answer

Use SageMaker Inference Recommender to find the optimal instance type and endpoint configuration.. SageMaker Inference Recommender runs load tests against the model to identify the optimal instance type, instance count, and endpoint configuration (e.g., container parameters, model server settings) that meet the specific latency and throughput requirements. For a large deep learning model demanding under 100 ms latency and 1000 requests per second, this automated benchmarking is essential to avoid over-provisioning or under-provisioning resources.

Answer

Deploy the model on a single endpoint with automatic scaling based on CPU utilization.

Answer

Use SageMaker Serverless Inference with provisioned concurrency.

Answer

Use a multi-model endpoint to load multiple copies of the model on the same instance.

Question 10

A company is building a recommendation system using Amazon SageMaker. The training data includes user-item interactions stored in a DataFrame with over 100 million rows. The data scientist wants to perform feature engineering, including one-hot encoding of categorical features with high cardinality. Which approach is MOST cost-effective and scalable?

Accepted Answer

Use SageMaker Processing with a Spark container to distribute the encoding job.. SageMaker Processing with a Spark container allows distributed execution of one-hot encoding on high-cardinality categorical features across a managed cluster, scaling horizontally to handle over 100 million rows without manual infrastructure management. This approach is cost-effective as you pay only for the processing time, and it integrates natively with SageMaker for seamless data pipeline orchestration.

Answer

Use Amazon EMR with Spark and store the processed data in HDFS.

Answer

Use a SageMaker notebook instance with scikit-learn to perform the encoding in memory.

Answer

Use AWS Glue ETL jobs to perform the encoding and store the result in S3.

Question 11

A machine learning team is using Amazon SageMaker Experiments to track multiple training runs. They need to compare the performance of different models based on metrics like accuracy and F1 score. However, when they view the experiment list in SageMaker Studio, the metrics are not displayed. What is the MOST likely cause?

Accepted Answer

The training script did not use the SageMaker SDK to log the metrics.. SageMaker Experiments automatically tracks parameters and metrics only when the training script explicitly logs them using the SageMaker SDK's `log_metric` function or the `sagemaker.experiments.run.Run` class. Without these SDK calls, the metrics are never recorded in the experiment's trial components, so they will not appear in the Studio experiment list. The team must instrument their training code to emit metrics for Experiments to capture them.

Answer

The training job did not define metric definitions in the algorithm specification.

Answer

The training job is running on an instance type that does not support Experiments.

Answer

The IAM role used by SageMaker does not have permission to write to the Experiments table.

Question 12

A company is deploying a PyTorch model on a SageMaker endpoint for real-time inference. The model is stored as a .pth file in an S3 bucket. The data scientist wants to use the SageMaker PyTorch inference toolkit. Which file is REQUIRED in the model artifacts to serve the model?

Accepted Answer

A file named model.pth containing the model state dictionary.. The SageMaker PyTorch inference toolkit expects the model artifact to be a single file named model.pth containing the model's state dictionary. When using the default inference handler, the toolkit automatically loads this file into the PyTorch model for serving. No additional inference script is required if the default behavior is sufficient.

Answer

A file named model.tar.gz that contains the model and any dependencies.

Answer

A file named inference.py that defines the model loading and prediction logic.

Answer

A file named requirements.txt listing the dependencies.

Question 13

A data scientist is using Amazon SageMaker Autopilot to automatically build a binary classification model. After the Autopilot job completes, the best model has an accuracy of 0.85 on the validation set. However, the data scientist notices a class imbalance (90% negative, 10% positive). Which metric should the data scientist use to evaluate the model's performance on the positive class?

Accepted Answer

Area Under the ROC Curve (AUC). A is correct because Area Under the ROC Curve (AUC) is threshold-independent and evaluates the model's ability to distinguish between positive and negative classes across all classification thresholds. In the presence of severe class imbalance (90% negative, 10% positive), AUC provides a robust measure of model performance on the positive class without being skewed by the majority class, unlike accuracy which would be high even if the model predicts all negatives.

Answer

Recall

Answer

Accuracy

Answer

Precision

Question 14

A data scientist is using Amazon SageMaker to train a model using a custom Docker container. The training job fails with an error message indicating that the container exited with a non-zero code. Which THREE steps should the data scientist take to diagnose the issue? (Choose THREE.)

Accepted Answer

Use the SageMaker Debugger to capture system metrics and output tensors for analysis.. Option B is correct because SageMaker Debugger captures system metrics (CPU, GPU, memory) and output tensors, which helps identify whether the non-zero exit was caused by resource exhaustion, vanishing gradients, or other runtime issues inside the custom container. Option C is correct because the container's stdout and stderr are streamed to CloudWatch Logs under the log group /aws/sagemaker/TrainingJobs, and inspecting those logs is the primary way to see the actual Python traceback or error that caused the non-zero exit code. Option E is correct because SageMaker Local Mode runs the same custom Docker container on the data scientist's local machine or notebook instance, allowing rapid iteration and direct debugging of the container's entrypoint, dependencies, and environment variables without waiting for a remote training job. Option A is not appropriate because retrying an identical configuration is unlikely to fix a deterministic container failure and wastes time and cost. Option D is not appropriate because adding training instances addresses scaling or distributed-training throughput, not a container that exits with a non-zero code due to a code, dependency, or configuration error.

Answer

Retry the training job with the same configuration; the error might be transient.

Answer

Increase the number of training instances to distribute the workload.

Question 15

A company runs a real-time fraud detection model on a SageMaker endpoint. The model is a TensorFlow neural network trained on transactional data. The endpoint uses a single ml.p3.2xlarge instance. Recently, the application’s latency has increased from 50ms to 500ms on average. The CloudWatch metrics show that CPU utilization is at 90%, GPU utilization is at 30%, and memory utilization is at 40%. The number of requests per second has remained stable. The ML team suspects the model is not fully utilizing the GPU. What action should the team take to reduce latency without changing the instance type?

Accepted Answer

Use SageMaker Neo to compile the model for the target instance. The CloudWatch metrics show CPU at 90% while GPU sits at only 30%, indicating the TensorFlow model is running in a CPU-bound mode rather than leveraging GPU acceleration. SageMaker Neo compiles the trained model into an optimized executable tuned for the target instance's specific hardware (including its GPU), reducing inference latency without requiring a larger instance. This directly addresses the bottleneck by shifting work off the CPU and onto the underutilized GPU.

Answer

Switch to SageMaker Batch Transform to process requests in batches

Answer

Change the endpoint to a compute-optimized instance like ml.c5.large

Answer

Increase the number of instances behind the endpoint and use a load balancer