MLS-C01 Machine Learning Implementation and Operations Practice Test 14 — 15 Questions

Question 1

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.. Option A is correct because a sudden spike in loss after decreasing often indicates a gradient explosion. Gradient clipping prevents this. Option B is wrong because learning rate that is too low causes slow convergence, not spikes. Option C is wrong because overfitting shows decreasing training loss but increasing validation loss. Option D is wrong because vanishing gradients cause loss to plateau, not spike.

Answer

Overfitting; apply regularization.

Answer

Learning rate too low; increase learning rate.

Answer

Vanishing gradients; use ReLU activation.

Question 2

A company wants to perform automated hyperparameter tuning for a model. Which Amazon SageMaker feature should be used?

Accepted Answer

Amazon SageMaker automatic model tuning. Option B is correct because SageMaker automatic model tuning (hyperparameter tuning) automates hyperparameter optimization. Option A is wrong because Debugger is for monitoring. Option C is wrong because Ground Truth is for labeling. Option D is wrong because Clarify is for bias detection.

Answer

Amazon SageMaker Clarify

Answer

Amazon SageMaker Ground Truth

Answer

Amazon SageMaker Debugger

Question 3

A machine learning engineer is building a pipeline using Amazon SageMaker Pipelines. The pipeline has multiple steps including data preprocessing, training, and evaluation. Which statement about SageMaker Pipelines is correct?

Accepted Answer

Pipelines support caching of step outputs.. Option D is correct because SageMaker Pipelines supports caching of step outputs to avoid re-execution. Option A is wrong because steps can be conditional. Option B is wrong because pipelines can include custom scripts. Option C is wrong because pipelines support parallel execution.

Answer

Steps in a pipeline must run sequentially.

Answer

Pipelines can only use built-in algorithms.

Answer

Pipelines cannot have conditional branches.

Question 4

A company is using Amazon SageMaker to deploy a model for real-time inference. The endpoint receives variable traffic and the company wants to optimize cost while maintaining responsiveness. Which scaling policy should be used?

Accepted Answer

Target tracking scaling based on invocation count. Option C is correct because target tracking scaling based on invocation count is the best approach for variable traffic, as it adjusts to demand. Option A is wrong because manual scaling requires constant monitoring. Option B is wrong because simple scaling with fixed steps may not adapt well. Option D is wrong because scheduled scaling is for predictable traffic.

Answer

Simple scaling with a cooldown period

Answer

Scheduled scaling

Answer

Manual scaling

Question 5

A data scientist needs to process a large dataset (100 TB) for training a machine learning model. The data is stored in Amazon S3. Which approach is most cost-effective and efficient for data processing?

Accepted Answer

Use Amazon EMR with Apache Spark.. Option A is correct because Amazon EMR with Spark is designed for large-scale data processing and is cost-effective. Option B is wrong because Athena is for querying, not complex processing. Option C is wrong because Glue is for ETL but may be slower for large data. Option D is wrong because SageMaker processing with a single instance may not handle 100 TB efficiently.

Answer

Use AWS Glue ETL jobs.

Answer

Use Amazon Athena to run SQL queries.

Answer

Use Amazon SageMaker Processing with a single large instance.

Question 6

A company is deploying a machine learning model using Amazon SageMaker. The model must be updated frequently without downtime. Which TWO strategies can achieve this? (Choose two.)

Accepted Answer

Use blue/green deployment with endpoint variants.. Options B and D are correct. Blue/green deployment and canary deployment both allow zero-downtime updates by routing traffic gradually. Option A is wrong because deleting and recreating the endpoint causes downtime. Option C is wrong because updating the model directly on the endpoint is not supported without creating a new endpoint. Option E is wrong because SageMaker does not support rolling updates natively.

Answer

Update the model artifact on the existing endpoint.

Answer

Delete the existing endpoint and create a new one.

Answer

Use rolling update with multiple instances.

Question 7

A data scientist is using Amazon SageMaker to train a model and wants to track experiments, including parameters and metrics. Which THREE actions should be taken? (Choose three.)

Accepted Answer

Create an experiment in SageMaker Experiments.. Options A, B, and D are correct. SageMaker Experiments tracks parameters and metrics; adding the SDK to the training script logs them; creating an experiment integrates tracking. Option C is wrong because CloudWatch is for monitoring but not designed for experiment tracking. Option E is wrong because SageMaker Studio is the interface, not the tracking mechanism itself.

Answer

Use SageMaker Studio to manually record experiments.

Answer

Use Amazon CloudWatch Logs to store experiment data.

Question 8

A machine learning engineer is setting up a training job in Amazon SageMaker. Which THREE components are required to define a training job? (Choose three.)

Accepted Answer

Output data configuration (e.g., model artifact path).. Options A, B, and D are correct. An algorithm or container, input data configuration, and output data configuration are required. Option C is wrong because hyperparameters are optional. Option E is wrong because an IAM role is required, not a VPC (though VPC is common).

Answer

VPC configuration for network isolation.

Answer

Hyperparameters for the algorithm.

Question 9

A company is training a deep learning model on Amazon SageMaker. The training job is failing with an out-of-memory error. Which SageMaker feature should the company use to resolve this issue without changing the instance type?

Accepted Answer

Use SageMaker distributed training with model parallelism. Option D is correct because SageMaker Managed Spot Training can reduce costs but does not fix memory issues. Option A and B are about debugging, not memory. Option C (distributed training) can split the model across instances, reducing per-instance memory usage. Option E is about cost, not memory.

Answer

Use SageMaker Savings Plans

Answer

Enable SageMaker Managed Spot Training

Answer

Use SageMaker Debugger to monitor memory usage

Answer

Enable SageMaker Profiler to profile memory

Question 10

A data scientist is deploying a machine learning model using SageMaker and wants to automate the retraining pipeline. The training data is updated daily in an S3 bucket. Which combination of AWS services should the data scientist use to trigger a new training job when new data arrives?

Accepted Answer

Amazon CloudWatch Events to invoke an AWS Lambda function that starts a SageMaker training job. Option C is correct because S3 Events can trigger a Lambda function, which starts a SageMaker training job. Option A uses CloudWatch Events but is more complex for simple triggers. Option B misses the Lambda function. Option D uses Step Functions unnecessarily. Option E uses SageMaker Pipelines, which can be triggered by events but requires additional setup.

Answer

Amazon SQS queue to store S3 events and a cron job to poll and start training

Answer

Use SageMaker Pipelines with a schedule to check for new data every hour

Answer

Amazon S3 event notification to directly start a SageMaker training job

Answer

Amazon CloudWatch Events to run an AWS Step Functions state machine that starts a SageMaker training job

Question 11

A company is using SageMaker to train a large NLP model. The training job is taking too long due to high I/O wait time. The data is stored as CSV files in S3. Which optimization should the company implement to reduce I/O wait time?

Accepted Answer

Use SageMaker Pipe mode to stream data directly from S3. Option B is correct because Pipe mode streams data directly from S3, reducing disk I/O. Option A (File mode) writes to disk first, causing I/O wait. Option C (ShardedByS3Key) is for distributed training but does not reduce I/O. Option D (RecordIO) reduces file size but still uses File mode. Option E (SageMaker batch transform) is for inference, not training.

Answer

Convert CSV files to RecordIO format

Answer

Use SageMaker batch transform before training

Answer

Use SageMaker File mode with larger instance storage

Answer

Use SageMaker ShardedByS3Key data distribution

Question 12

A machine learning team is using SageMaker to build a model. They need to track hyperparameter tuning experiments, compare results, and visualize metrics. Which SageMaker feature should they use?

Accepted Answer

SageMaker Experiments. Option C is correct because SageMaker Experiments provides experiment tracking, comparison, and visualization. Option A (Hyperparameter Tuning) only tunes, not tracks. Option B (Debugger) is for debugging. Option D (Model Monitor) is for monitoring after deployment. Option E (Ground Truth) is for labeling.

Answer

SageMaker Ground Truth

Answer

SageMaker Model Monitor

Answer

SageMaker Hyperparameter Tuning

Answer

SageMaker Debugger

Question 13

A company has deployed a model on SageMaker for real-time inference. The endpoint is experiencing high latency during traffic spikes. Which action should the company take to reduce latency?

Accepted Answer

Enable SageMaker endpoint auto-scaling. Option D is correct because enabling auto-scaling adds instances during spikes, reducing latency. Option A (larger instance) may help but is not cost-effective. Option B (batch transform) is for async inference. Option C (Elastic Inference) accelerates inference but does not handle spikes. Option E (SageMaker Neo) optimizes for edge devices.

Answer

Use a larger instance type for the endpoint

Answer

Attach SageMaker Elastic Inference to the endpoint

Answer

Use SageMaker Neo to compile the model

Answer

Switch to SageMaker batch transform

Question 14

A data scientist is using SageMaker to train a model with a custom algorithm. The training script uses TensorFlow and runs on GPU instances. The training job fails with 'CUDA_ERROR_OUT_OF_MEMORY'. What is the most likely cause?

Accepted Answer

The batch size is too large for the GPU memory. Option C is correct because the error indicates GPU memory exhaustion, often due to batch size being too large. Option A (insufficient CPU) would not cause CUDA errors. Option B (S3 bandwidth) is unrelated. Option D (CPU memory) is not the issue. Option E (driver version) would give a different error.

Answer

The S3 bucket is in a different region

Answer

The GPU driver is outdated

Answer

The training script has a memory leak on CPU

Answer

The instance type does not have enough CPU cores

Question 15

A company wants to use SageMaker to host multiple models behind a single endpoint to reduce costs. Which SageMaker feature should they use?

Accepted Answer

SageMaker Multi-Model Endpoints. Option C is correct because SageMaker Multi-Model Endpoints allow multiple models on the same endpoint. Option A (multi-container) is for different containers, not multiple models. Option B (batch transform) is offline. Option D (inference pipeline) chains containers. Option E (Elastic Inference) accelerates inference.

Answer

SageMaker Elastic Inference

Answer

SageMaker inference pipeline

Answer

SageMaker batch transform

Answer

SageMaker multi-container endpoints