MLS-C01 Machine Learning Implementation and Operations Practice Test 7 — 15 Questions

Question 1

A machine learning engineer is deploying a TensorFlow model to an Amazon SageMaker endpoint. The endpoint is behind an Application Load Balancer (ALB) for A/B testing. The engineer notices that the new variant is not receiving any traffic. What is the most likely cause?

Accepted Answer

The ALB target group weight for the new variant is set to 0.. Option A is correct because the ALB target group weight determines traffic distribution. Option B is wrong because routing is based on weights, not health checks if healthy. Option C is wrong because SageMaker endpoint is the target. Option D is wrong because it would cause errors, not zero traffic.

Answer

The new variant's health checks are failing.

Answer

The model is not compatible with the ALB's protocol.

Answer

The ALB is not configured to route to SageMaker endpoints.

Question 2

A data scientist is using Amazon SageMaker to train a model. The training data is stored in an S3 bucket encrypted with AWS KMS. Which TWO actions are necessary to allow SageMaker to access the data?

Accepted Answer

Ensure the SageMaker execution role has s3:GetObject permission.. Option A is correct because SageMaker needs permission to decrypt. Option C is correct because the execution role needs permission. Option B is wrong because bucket policy is separate. Option D is wrong because it's not required. Option E is wrong because VPC endpoint is not required.

Answer

Enable S3 Transfer Acceleration.

Answer

Set up a VPC endpoint for S3.

Answer

Add a bucket policy allowing SageMaker access.

Question 3

A company is deploying a machine learning model using Amazon SageMaker. The model needs to be updated frequently. Which THREE practices should the company implement for model versioning and deployment?

Accepted Answer

Use AWS CodePipeline to automate the training and deployment pipeline.. Option A is correct for versioning. Option C is correct for canary deployments. Option E is correct for automated pipelines. Option B is wrong because it's for data. Option D is wrong because instances should be monitored.

Answer

Manually update the endpoint configuration each time.

Answer

Store all training datasets in a single S3 bucket without versioning.

Question 4

A company is using Amazon SageMaker to build a custom model. The training job is failing with a 'ResourceLimitExceeded' error. Which TWO actions should the company take to resolve this issue?

Accepted Answer

Request a service quota increase for SageMaker training instances.. Option A is correct because it increases the limit. Option D is correct because spot instances can help. Option B is wrong because it doesn't address limit. Option C is wrong because it's for storage. Option E is wrong because it doesn't help with compute limits.

Answer

Use Amazon EFS for training data.

Answer

Reduce the size of the training dataset.

Answer

Use a smaller instance type.

Question 5

A data scientist is trying to create a SageMaker training job using an execution role with the attached IAM policy. The training job fails with an access denied error when trying to read training data from the S3 bucket 'my-bucket'. What is the most likely cause?

Accepted Answer

The S3 bucket policy explicitly denies access to the role.. Option D is correct because the bucket policy may deny access. Option A is wrong because the role has s3:GetObject. Option B is wrong because it's allowed. Option C is wrong because it's allowed.

Answer

The IAM policy does not include s3:ListBucket permission.

Answer

The S3 bucket is in a different AWS account.

Answer

The sagemaker:CreateTrainingJob action is not allowed.

Question 6

A data scientist is reviewing the training logs from a SageMaker training job. The model's loss decreases steadily and accuracy increases. However, when the model is evaluated on a holdout test set, the accuracy is only 0.65. Which issue does this behavior suggest?

Accepted Answer

The model is overfitting to the training data.. Option C is correct because low test accuracy indicates overfitting. Option A is wrong because loss and accuracy improve. Option B is wrong because it's not a data issue. Option D is wrong because it's not underfitting.

Answer

The learning rate is too high.

Answer

The model is underfitting the training data.

Answer

The training data is corrupted.

Question 7

A data scientist submits a SageMaker training job with the provided configuration. The job fails immediately with the error 'Algorithm not found: 382416733822.dkr.ecr.us-west-2.amazonaws.com/sagemaker-xgboost:1.2-1'. What is the most likely cause?

Accepted Answer

The ECR repository URI is incorrect or the image does not exist.. Option A is correct because the image URI may be wrong. Option B is wrong because region is specified. Option C is wrong because role is specified. Option D is wrong because format is correct.

Answer

The training region is different from the image region.

Answer

The input data format is incorrect.

Answer

The IAM role does not have permission to pull the image.

Question 8

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 captures internal model state such as gradients and tensors during training, enabling analysis and debugging. SageMaker Profiler (B) focuses on system performance, not model internals. SageMaker Experiments (C) tracks trials and metrics but does not capture internal state. SageMaker Model Monitor (D) detects data drift after deployment.

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 9

A company deploys a real-time inference endpoint using Amazon SageMaker with an ML model that has strict latency requirements. The endpoint currently uses a single ml.c5.xlarge instance. During a load test, the p99 latency exceeds the 100ms threshold. The team adds more instances but latency does not improve because the model is heavily CPU-bound. What is the MOST cost-effective change to meet the latency requirement?

Accepted Answer

Change the instance type to a GPU instance such as ml.g4dn.xlarge.. Switching to an instance with GPU acceleration (e.g., ml.g4dn.xlarge) offloads computation to GPU, reducing CPU-bound latency. More instances (B) increase throughput but not per-request latency if the model is CPU-bound. Multi-model endpoints (C) help with many models but not single-model latency. Automatic scaling (D) helps with varying load but not per-request latency improvement.

Answer

Use a multi-model endpoint to serve multiple models on the same instance.

Answer

Enable automatic scaling based on inference latency.

Answer

Increase the number of instances and use a target tracking scaling policy.

Question 10

An ML engineer needs to run a hyperparameter tuning job on Amazon SageMaker. The training algorithm supports distributed training across multiple GPUs. The engineer wants to minimize the total time to find the best hyperparameters. Which strategy should be used?

Accepted Answer

Use Bayesian optimization as the tuning strategy.. Bayesian search uses past results to select hyperparameters, converging faster than random or grid search. Random search (B) does not use past results. Grid search (C) is exhaustive and slow. Hyperband (D) is a bandit method but requires early stopping; Bayesian is better for minimizing total time with a fixed budget.

Answer

Use random search to explore a wide range.

Answer

Use grid search to cover all combinations.

Answer

Use Hyperband which is designed for distributed training.

Question 11

A company is deploying a machine learning model for real-time fraud detection using Amazon SageMaker. The model must have a p99 inference latency under 50ms. Which TWO actions should the ML team take to meet the latency requirement?

Accepted Answer

Use SageMaker Neo to compile and optimize the model for the target instance type.. Using SageMaker Neo (A) optimizes the model for the target hardware, reducing latency. Using GPU instances (B) can accelerate inference for compute-intensive models. SageMaker Batch Transform (C) is for offline inference. Automatic scaling (D) handles throughput, not latency. Multi-model endpoint (E) can help with many models but not single-model latency.

Answer

Use a multi-model endpoint to reduce cold starts.

Answer

Use SageMaker Batch Transform for near-real-time inference.

Answer

Configure automatic scaling to add instances based on CPU utilization.

Question 12

A machine learning team is building a real-time inference pipeline using Amazon SageMaker. The team has multiple models that need to be served, but usage patterns are unpredictable and traffic spikes occur several times a day. The team wants to minimize costs while maintaining low latency. Which THREE actions should the team take?

Accepted Answer

Use SageMaker inference with Spot Instances to reduce cost.. SageMaker multi-model endpoints (A) allow serving multiple models on a single instance, reducing cost. SageMaker automatic scaling (B) adjusts capacity based on demand, handling spikes. Using Spot Instances (C) for inference can reduce cost but may cause interruptions; for real-time, On-Demand is safer. Provisioned concurrency (D) is for Lambda, not SageMaker. Batch Transform (E) is for offline inference.

Answer

Enable provisioned concurrency on the endpoint to reduce cold starts.

Answer

Use SageMaker Batch Transform for all inference requests.

Question 13

A data scientist is using Amazon SageMaker to train a large neural network on a GPU instance. The training is taking longer than expected. The scientist wants to reduce training time without changing the model architecture. Which TWO approaches should the scientist consider?

Accepted Answer

Use SageMaker's distributed training with multiple GPU instances.. Using multiple GPU instances with SageMaker distributed training (A) can accelerate training. Using SageMaker Managed Spot Training (B) reduces cost but not time. Using SageMaker Debugger (C) helps debugging but not speed. SageMaker Automatic Model Tuning (D) is for hyperparameter optimization. Using a larger GPU instance (E) with more memory and compute can directly reduce training time.

Answer

Use SageMaker Automatic Model Tuning to find optimal hyperparameters.

Answer

Use SageMaker Managed Spot Training to reduce cost.

Answer

Enable SageMaker Debugger to capture training metrics.

Question 14

An ML engineer is deploying a model on a SageMaker endpoint and wants to ensure that only authorized users and services can invoke the endpoint. The company uses AWS IAM for access control and requires that the endpoint be invoked only from within a specific VPC. What combination of actions should the engineer take? (Choose the single best answer.)

Accepted Answer

Create a VPC endpoint for Amazon SageMaker and attach a policy that only allows invocation from the VPC. Use IAM roles to restrict which users can invoke the endpoint.. The best approach is to attach an IAM policy to the endpoint (via AWS Lambda or resource policy) that denies access unless the request originates from the VPC. However, SageMaker endpoints do not support resource-based policies directly; instead, use VPC Endpoint policies and IAM. The correct answer is to create a VPC endpoint for SageMaker and attach a policy that restricts invocation to the VPC, combined with IAM roles that allow only specific principals. Option B (network ACLs) is not sufficient for authentication. Option C (API Gateway) adds unnecessary complexity. Option D (CloudFront) is for CDN, not access control.

Answer

Use AWS CloudFront to restrict access based on IP addresses.

Answer

Use API Gateway in front of the SageMaker endpoint and attach a resource policy to API Gateway.

Answer

Configure network ACLs on the VPC subnet to allow only the endpoint's security group.

Question 15

A company uses Amazon SageMaker to train a model using a custom Docker container. The training job fails with an error: "Unable to write to /opt/ml/output/data". The data scientist checks the container and finds that the /opt/ml directory is not writable. What is the MOST likely cause?

Accepted Answer

The container runs as a non-root user that lacks write permissions to /opt/ml.. SageMaker expects the training container to write output to /opt/ml/model and /opt/ml/output. If the container does not have write permissions to these directories, the training fails. The most likely cause is that the Docker image was built with a non-root user that does not have write permissions to /opt/ml. The correct fix is to ensure the container runs as root or grants write permissions.

Answer

The Docker image is built from a base image that does not have the required libraries.

Answer

The SageMaker training job is configured with insufficient memory.

Answer

The training script is not copying the model to /opt/ml/model.