Amazon MLS-C01 Practice Exams
Last updated on Oct 23,2026- Exam Code: MLS-C01
- Exam Name: AWS Certified Machine Learning - Specialty
- Certification Provider: Amazon
- Latest update: Oct 23,2026
Question #291
A company’s data scientist has trained a new machine learning model that performs better on test data than the company’s existing model performs in the production environment. The data scientist wants to replace the existing model that runs on an Amazon SageMaker endpoint in the production environment. However, the company is concerned that the new model might not work well on the production environment data.
The data scientist needs to perform A/B testing in the production environment to evaluate whether the new model performs well on production environment data.
Which combination of steps must the data scientist take to perform the A/B testing? (Choose two.)
- A . Create a new endpoint configuration that includes a production variant for each of the two models.
- B . Create a new endpoint configuration that includes two target variants that point to different endpoints.
- C . Deploy the new model to the existing endpoint.
- D . Update the existing endpoint to activate the new model.
- E . Update the existing endpoint to use the new endpoint configuration.
Correct Answer: A, E
A, E
Explanation:
The combination of steps that the data scientist must take to perform the A/B testing are to create a new endpoint configuration that includes a production variant for each of the two models, and update the existing endpoint to use the new endpoint configuration. This approach will allow the data scientist to deploy both models on the same endpoint and split the inference traffic between them based on a specified distribution.
Amazon SageMaker is a fully managed service that provides developers and data scientists the ability to quickly build, train, and deploy machine learning models. Amazon SageMaker supports A/B testing on machine learning models by allowing the data scientist to run multiple production variants on an endpoint. A production variant is a version of a model that is deployed on an endpoint. Each production variant has a name, a machine learning model, an instance type, an initial instance count, and an initial weight. The initial weight determines the percentage of inference requests that the variant will handle. For example, if there are two variants with weights of 0.5 and 0.5, each variant will handle 50% of the requests. The data scientist can use production variants to test models that have been trained using different training datasets, algorithms, and machine learning frameworks; test how they perform on different instance types; or a combination of all of the above1.
To perform A/B testing on machine learning models, the data scientist needs to create a new endpoint configuration that includes a production variant for each of the two models. An endpoint configuration is a collection of settings that define the properties of an endpoint, such as the name, the production variants, and the data capture configuration. The data scientist can use the Amazon SageMaker console, the AWS CLI, or the AWS SDKs to create a new endpoint configuration. The data scientist needs to specify the name, model name, instance type, initial instance count, and initial variant weight for each production variant in the endpoint configuration2.
After creating the new endpoint configuration, the data scientist needs to update the existing endpoint to use the new endpoint configuration. Updating an endpoint is the process of deploying a new endpoint configuration to an existing endpoint. Updating an endpoint does not affect the availability or scalability of the endpoint, as Amazon SageMaker creates a new endpoint instance with the new configuration and switches the DNS record to point to the new instance when it is ready. The data scientist can use the Amazon SageMaker console, the AWS CLI, or the AWS SDKs to update an endpoint. The data scientist needs to specify the name of the endpoint and the name of the new endpoint configuration to update the endpoint3.
The other options are either incorrect or unnecessary. Creating a new endpoint configuration that includes two target variants that point to different endpoints is not possible, as target variants are only used to invoke a specific variant on an endpoint, not to define an endpoint configuration.
Deploying the new model to the existing endpoint would replace the existing model, not run it side-by-side with the new model. Updating the existing endpoint to activate the new model is not a valid operation, as there is no activation parameter for an endpoint.
References:
1: A/B Testing ML models in production using Amazon SageMaker | AWS Machine Learning Blog
2: Create an Endpoint Configuration – Amazon SageMaker
3: Update an Endpoint – Amazon SageMaker
A, E
Explanation:
The combination of steps that the data scientist must take to perform the A/B testing are to create a new endpoint configuration that includes a production variant for each of the two models, and update the existing endpoint to use the new endpoint configuration. This approach will allow the data scientist to deploy both models on the same endpoint and split the inference traffic between them based on a specified distribution.
Amazon SageMaker is a fully managed service that provides developers and data scientists the ability to quickly build, train, and deploy machine learning models. Amazon SageMaker supports A/B testing on machine learning models by allowing the data scientist to run multiple production variants on an endpoint. A production variant is a version of a model that is deployed on an endpoint. Each production variant has a name, a machine learning model, an instance type, an initial instance count, and an initial weight. The initial weight determines the percentage of inference requests that the variant will handle. For example, if there are two variants with weights of 0.5 and 0.5, each variant will handle 50% of the requests. The data scientist can use production variants to test models that have been trained using different training datasets, algorithms, and machine learning frameworks; test how they perform on different instance types; or a combination of all of the above1.
To perform A/B testing on machine learning models, the data scientist needs to create a new endpoint configuration that includes a production variant for each of the two models. An endpoint configuration is a collection of settings that define the properties of an endpoint, such as the name, the production variants, and the data capture configuration. The data scientist can use the Amazon SageMaker console, the AWS CLI, or the AWS SDKs to create a new endpoint configuration. The data scientist needs to specify the name, model name, instance type, initial instance count, and initial variant weight for each production variant in the endpoint configuration2.
After creating the new endpoint configuration, the data scientist needs to update the existing endpoint to use the new endpoint configuration. Updating an endpoint is the process of deploying a new endpoint configuration to an existing endpoint. Updating an endpoint does not affect the availability or scalability of the endpoint, as Amazon SageMaker creates a new endpoint instance with the new configuration and switches the DNS record to point to the new instance when it is ready. The data scientist can use the Amazon SageMaker console, the AWS CLI, or the AWS SDKs to update an endpoint. The data scientist needs to specify the name of the endpoint and the name of the new endpoint configuration to update the endpoint3.
The other options are either incorrect or unnecessary. Creating a new endpoint configuration that includes two target variants that point to different endpoints is not possible, as target variants are only used to invoke a specific variant on an endpoint, not to define an endpoint configuration.
Deploying the new model to the existing endpoint would replace the existing model, not run it side-by-side with the new model. Updating the existing endpoint to activate the new model is not a valid operation, as there is no activation parameter for an endpoint.
References:
1: A/B Testing ML models in production using Amazon SageMaker | AWS Machine Learning Blog
2: Create an Endpoint Configuration – Amazon SageMaker
3: Update an Endpoint – Amazon SageMaker
Question #292
A company builds computer-vision models that use deep learning for the autonomous vehicle industry. A machine learning (ML) specialist uses an Amazon EC2 instance that has a CPU: GPU ratio of 12:1 to train the models.
The ML specialist examines the instance metric logs and notices that the GPU is idle half of the time
The ML specialist must reduce training costs without increasing the duration of the training jobs.
Which solution will meet these requirements?
- A . Switch to an instance type that has only CPUs.
- B . Use a heterogeneous cluster that has two different instances groups.
- C . Use memory-optimized EC2 Spot Instances for the training jobs.
- D . Switch to an instance type that has a CPU GPU ratio of 6:1.
Correct Answer: D
D
Explanation:
Switching to an instance type that has a CPU: GPU ratio of 6:1 will reduce the training costs by using fewer CPUs and GPUs, while maintaining the same level of performance. The GPU idle time indicates that the CPU is not able to feed the GPU with enough data, so reducing the CPU: GPU ratio will balance the workload and improve the GPU utilization. A lower CPU: GPU ratio also means less overhead for inter-process communication and synchronization between the CPU and GPU processes.
References:
Optimizing GPU utilization for AI/ML workloads on Amazon EC2 Analyze CPU vs. GPU Performance for AWS Machine Learning
D
Explanation:
Switching to an instance type that has a CPU: GPU ratio of 6:1 will reduce the training costs by using fewer CPUs and GPUs, while maintaining the same level of performance. The GPU idle time indicates that the CPU is not able to feed the GPU with enough data, so reducing the CPU: GPU ratio will balance the workload and improve the GPU utilization. A lower CPU: GPU ratio also means less overhead for inter-process communication and synchronization between the CPU and GPU processes.
References:
Optimizing GPU utilization for AI/ML workloads on Amazon EC2 Analyze CPU vs. GPU Performance for AWS Machine Learning
Question #293
A Machine Learning Specialist is assigned a TensorFlow project using Amazon SageMaker for training, and needs to continue working for an extended period with no Wi-Fi access.
Which approach should the Specialist use to continue working?
- A . Install Python 3 and boto3 on their laptop and continue the code development using that environment.
- B . Download the TensorFlow Docker container used in Amazon SageMaker from GitHub to their local environment, and use the Amazon SageMaker Python SDK to test the code.
- C . Download TensorFlow from tensorflow.org to emulate the TensorFlow kernel in the SageMaker environment.
- D . Download the SageMaker notebook to their local environment then install Jupyter Notebooks on their laptop and continue the development in a local notebook.
Correct Answer: B
B
Explanation:
Amazon SageMaker is a fully managed service that enables developers and data scientists to quickly and easily build, train, and deploy machine learning models at any scale. SageMaker provides a variety of tools and frameworks to support the entire machine learning workflow, from data preparation to model deployment.
One of the tools that SageMaker offers is the Amazon SageMaker Python SDK, which is a high-level library that simplifies the interaction with SageMaker APIs and services. The SageMaker Python SDK allows you to write code in Python and use popular frameworks such as TensorFlow, PyTorch, MXNet, and more. You can use the SageMaker Python SDK to create and manage SageMaker resources such as notebook instances, training jobs, endpoints, and feature store.
If you need to continue working on a TensorFlow project using SageMaker for training without Wi-Fi access, the best approach is to download the TensorFlow Docker container used in SageMaker from GitHub to your local environment, and use the SageMaker Python SDK to test the code. This way, you can ensure that your code is compatible with the SageMaker environment and avoid any potential issues when you upload your code to SageMaker and start the training job. You can also use the same code to deploy your model to a SageMaker endpoint when you have Wi-Fi access again.
To download the TensorFlow Docker container used in SageMaker, you can visit the SageMaker Docker GitHub repository and follow the instructions to build the image locally. You can also use the SageMaker Studio Image Build CLI to automate the process of building and pushing the Docker image to Amazon Elastic Container Registry (Amazon ECR). To use the SageMaker Python SDK to test the code, you can install the SDK on your local machine by following the installation guide. You can also refer to the TensorFlow documentation for more details on how to use the SageMaker Python SDK with TensorFlow.
References:
SageMaker Docker GitHub repository
SageMaker Studio Image Build CLI
SageMaker Python SDK installation guide
SageMaker Python SDK TensorFlow documentation
B
Explanation:
Amazon SageMaker is a fully managed service that enables developers and data scientists to quickly and easily build, train, and deploy machine learning models at any scale. SageMaker provides a variety of tools and frameworks to support the entire machine learning workflow, from data preparation to model deployment.
One of the tools that SageMaker offers is the Amazon SageMaker Python SDK, which is a high-level library that simplifies the interaction with SageMaker APIs and services. The SageMaker Python SDK allows you to write code in Python and use popular frameworks such as TensorFlow, PyTorch, MXNet, and more. You can use the SageMaker Python SDK to create and manage SageMaker resources such as notebook instances, training jobs, endpoints, and feature store.
If you need to continue working on a TensorFlow project using SageMaker for training without Wi-Fi access, the best approach is to download the TensorFlow Docker container used in SageMaker from GitHub to your local environment, and use the SageMaker Python SDK to test the code. This way, you can ensure that your code is compatible with the SageMaker environment and avoid any potential issues when you upload your code to SageMaker and start the training job. You can also use the same code to deploy your model to a SageMaker endpoint when you have Wi-Fi access again.
To download the TensorFlow Docker container used in SageMaker, you can visit the SageMaker Docker GitHub repository and follow the instructions to build the image locally. You can also use the SageMaker Studio Image Build CLI to automate the process of building and pushing the Docker image to Amazon Elastic Container Registry (Amazon ECR). To use the SageMaker Python SDK to test the code, you can install the SDK on your local machine by following the installation guide. You can also refer to the TensorFlow documentation for more details on how to use the SageMaker Python SDK with TensorFlow.
References:
SageMaker Docker GitHub repository
SageMaker Studio Image Build CLI
SageMaker Python SDK installation guide
SageMaker Python SDK TensorFlow documentation
Question #294
A law firm handles thousands of contracts every day. Every contract must be signed. Currently, a lawyer manually checks all contracts for signatures.
The law firm is developing a machine learning (ML) solution to automate signature detection for each contract. The ML solution must also provide a confidence score for each contract page.
Which Amazon Textract API action can the law firm use to generate a confidence score for each page of each contract?
- A . Use the AnalyzeDocument API action. Set the FeatureTypes parameter to SIGNATURES. Return the
confidence scores for each page. - B . Use the Prediction API call on the documents. Return the signatures and confidence scores for each page.
- C . Use the StartDocumentAnalysis API action to detect the signatures. Return the confidence scores for each page.
- D . Use the GetDocumentAnalysis API action to detect the signatures. Return the confidence scores for each page
Correct Answer: A
A
The AnalyzeDocument API action is the best option to generate a confidence score for each page of each contract. This API action analyzes an input document for relationships between detected items. The input document can be an image file in JPEG or PNG format, or a PDF file. The output is a JSON structure that contains the extracted data from the document. The FeatureTypes parameter specifies the types of analysis to perform on the document. The available feature types are TABLES, FORMS, and SIGNATURES. By setting the FeatureTypes parameter to SIGNATURES, the API action will detect and extract information about signatures from the document. The output will include a list of SignatureDetection objects, each containing information about a detected signature, such as its location and confidence score. The confidence score is a value between 0 and 100 that indicates the probability that the detected signature is correct. The output will also include a list of Block objects, each representing a document page. Each Block object will have a Page attribute that contains the page number and a Confidence attribute that contains the confidence score for the page. The confidence score for the page is the average of the confidence scores of the blocks that are detected on the page. The law firm can use the AnalyzeDocument API action to generate a confidence score for each page of each contract by using the SIGNATURES feature type and returning the confidence scores from the SignatureDetection and Block objects.
The other options are not suitable for generating a confidence score for each page of each contract.
The Prediction API call is not an Amazon Textract API action, but a generic term for making inference requests to a machine learning model. The StartDocumentAnalysis API action is used to start an asynchronous job to analyze a document. The output is a job identifier (JobId) that is used to get the results of the analysis with the GetDocumentAnalysis API action. The GetDocumentAnalysis API action is used to get the results of a document analysis started by the StartDocumentAnalysis API action. The output is a JSON structure that contains the extracted data from the document. However, both the StartDocumentAnalysis and the GetDocumentAnalysis API actions do not support the SIGNATURES feature type, and therefore cannot detect signatures or provide confidence scores for them.
References:
• AnalyzeDocument
• SignatureDetection
• Block
• Amazon Textract launches the ability to detect signatures on any document
A
The AnalyzeDocument API action is the best option to generate a confidence score for each page of each contract. This API action analyzes an input document for relationships between detected items. The input document can be an image file in JPEG or PNG format, or a PDF file. The output is a JSON structure that contains the extracted data from the document. The FeatureTypes parameter specifies the types of analysis to perform on the document. The available feature types are TABLES, FORMS, and SIGNATURES. By setting the FeatureTypes parameter to SIGNATURES, the API action will detect and extract information about signatures from the document. The output will include a list of SignatureDetection objects, each containing information about a detected signature, such as its location and confidence score. The confidence score is a value between 0 and 100 that indicates the probability that the detected signature is correct. The output will also include a list of Block objects, each representing a document page. Each Block object will have a Page attribute that contains the page number and a Confidence attribute that contains the confidence score for the page. The confidence score for the page is the average of the confidence scores of the blocks that are detected on the page. The law firm can use the AnalyzeDocument API action to generate a confidence score for each page of each contract by using the SIGNATURES feature type and returning the confidence scores from the SignatureDetection and Block objects.
The other options are not suitable for generating a confidence score for each page of each contract.
The Prediction API call is not an Amazon Textract API action, but a generic term for making inference requests to a machine learning model. The StartDocumentAnalysis API action is used to start an asynchronous job to analyze a document. The output is a job identifier (JobId) that is used to get the results of the analysis with the GetDocumentAnalysis API action. The GetDocumentAnalysis API action is used to get the results of a document analysis started by the StartDocumentAnalysis API action. The output is a JSON structure that contains the extracted data from the document. However, both the StartDocumentAnalysis and the GetDocumentAnalysis API actions do not support the SIGNATURES feature type, and therefore cannot detect signatures or provide confidence scores for them.
References:
• AnalyzeDocument
• SignatureDetection
• Block
• Amazon Textract launches the ability to detect signatures on any document
Question #295
A finance company needs to forecast the price of a commodity. The company has compiled a dataset of historical daily prices. A data scientist must train various forecasting models on 80% of the dataset and must validate the efficacy of those models on the remaining 20% of the dataset.
What should the data scientist split the dataset into a training dataset and a validation dataset to compare model performance?
- A . Pick a date so that 80% to the data points precede the date Assign that group of data points as the training dataset. Assign all the remaining data points to the validation dataset.
- B . Pick a date so that 80% of the data points occur after the date. Assign that group of data points as the training dataset. Assign all the remaining data points to the validation dataset.
- C . Starting from the earliest date in the dataset. pick eight data points for the training dataset and two data points for the validation dataset. Repeat this stratified sampling until no data points remain.
- D . Sample data points randomly without replacement so that 80% of the data points are in the training dataset. Assign all the remaining data points to the validation dataset.
Correct Answer: A
A
Explanation:
A Comprehensive Explanation
The best way to split the dataset into a training dataset and a validation dataset is to pick a date so that 80% of the data points precede the date and assign that group of data points as the training dataset. This method preserves the temporal order of the data and ensures that the validation dataset reflects the most recent trends and patterns in the commodity price. This is important for forecasting models that rely on time series analysis and sequential data. The other methods would either introduce bias or lose information by ignoring the temporal structure of the data.
References:
Time Series Forecasting – Amazon SageMaker
Time Series Splitting – scikit-learn
Time Series Forecasting – Towards Data Science
A
Explanation:
A Comprehensive Explanation
The best way to split the dataset into a training dataset and a validation dataset is to pick a date so that 80% of the data points precede the date and assign that group of data points as the training dataset. This method preserves the temporal order of the data and ensures that the validation dataset reflects the most recent trends and patterns in the commodity price. This is important for forecasting models that rely on time series analysis and sequential data. The other methods would either introduce bias or lose information by ignoring the temporal structure of the data.
References:
Time Series Forecasting – Amazon SageMaker
Time Series Splitting – scikit-learn
Time Series Forecasting – Towards Data Science
Question #296
A developer at a retail company is creating a daily demand forecasting model. The company stores the historical hourly demand data in an Amazon S3 bucket. However, the historical data does not include demand data for some hours.
The developer wants to verify that an autoregressive integrated moving average (ARIMA) approach will be a suitable model for the use case.
How should the developer verify the suitability of an ARIMA approach?
- A . Use Amazon SageMaker Data Wrangler. Import the data from Amazon S3. Impute hourly missing data. Perform a Seasonal Trend decomposition.
- B . Use Amazon SageMaker Autopilot. Create a new experiment that specifies the S3 data location.
Choose ARIMA as the machine learning (ML) problem. Check the model performance. - C . Use Amazon SageMaker Data Wrangler. Import the data from Amazon S3. Resample data by using the aggregate daily total. Perform a Seasonal Trend decomposition.
- D . Use Amazon SageMaker Autopilot. Create a new experiment that specifies the S3 data location. Impute missing hourly values. Choose ARIMA as the machine learning (ML) problem. Check the model performance.
Correct Answer: A
A
Explanation:
The best solution to verify the suitability of an ARIMA approach is to use Amazon SageMaker Data Wrangler. Data Wrangler is a feature of SageMaker Studio that provides an end-to-end solution for importing, preparing, transforming, featurizing, and analyzing data. Data Wrangler includes built-in analyses that help generate visualizations and data insights in a few clicks. One of the built-in analyses is the Seasonal-Trend decomposition, which can be used to decompose a time series into its trend, seasonal, and residual components. This analysis can help the developer understand the patterns and characteristics of the time series, such as stationarity, seasonality, and autocorrelation, which are important for choosing an appropriate ARIMA model. Data Wrangler also provides built-in transformations that can help the developer handle missing data, such as imputing with mean, median, mode, or constant values, or dropping rows with missing values. Imputing missing data can help avoid gaps and irregularities in the time series, which can affect the ARIMA model performance. Data Wrangler also allows the developer to export the prepared data and the analysis code to various destinations, such as SageMaker Processing, SageMaker Pipelines, or SageMaker Feature Store, for further processing and modeling.
The other options are not suitable for verifying the suitability of an ARIMA approach. Amazon SageMaker Autopilot is a feature-set that automates key tasks of an automatic machine learning (AutoML) process. It explores the data, selects the algorithms relevant to the problem type, and prepares the data to facilitate model training and tuning. However, Autopilot does not support ARIMA as a machine learning problem type, and it does not provide any visualization or analysis of the time series data. Resampling data by using the aggregate daily total can reduce the granularity and resolution of the time series, which can affect the ARIMA model accuracy and applicability.
References:
• Analyze and Visualize
• Transform and Export
• Amazon SageMaker Autopilot
• ARIMA Model C Complete Guide to Time Series Forecasting in Python
A
Explanation:
The best solution to verify the suitability of an ARIMA approach is to use Amazon SageMaker Data Wrangler. Data Wrangler is a feature of SageMaker Studio that provides an end-to-end solution for importing, preparing, transforming, featurizing, and analyzing data. Data Wrangler includes built-in analyses that help generate visualizations and data insights in a few clicks. One of the built-in analyses is the Seasonal-Trend decomposition, which can be used to decompose a time series into its trend, seasonal, and residual components. This analysis can help the developer understand the patterns and characteristics of the time series, such as stationarity, seasonality, and autocorrelation, which are important for choosing an appropriate ARIMA model. Data Wrangler also provides built-in transformations that can help the developer handle missing data, such as imputing with mean, median, mode, or constant values, or dropping rows with missing values. Imputing missing data can help avoid gaps and irregularities in the time series, which can affect the ARIMA model performance. Data Wrangler also allows the developer to export the prepared data and the analysis code to various destinations, such as SageMaker Processing, SageMaker Pipelines, or SageMaker Feature Store, for further processing and modeling.
The other options are not suitable for verifying the suitability of an ARIMA approach. Amazon SageMaker Autopilot is a feature-set that automates key tasks of an automatic machine learning (AutoML) process. It explores the data, selects the algorithms relevant to the problem type, and prepares the data to facilitate model training and tuning. However, Autopilot does not support ARIMA as a machine learning problem type, and it does not provide any visualization or analysis of the time series data. Resampling data by using the aggregate daily total can reduce the granularity and resolution of the time series, which can affect the ARIMA model accuracy and applicability.
References:
• Analyze and Visualize
• Transform and Export
• Amazon SageMaker Autopilot
• ARIMA Model C Complete Guide to Time Series Forecasting in Python
Question #297
A power company wants to forecast future energy consumption for its customers in residential properties and commercial business properties. Historical power consumption data for the last 10 years is available. A team of data scientists who performed the initial data analysis and feature selection will include the historical power consumption data and data such as weather, number of individuals on the property, and public holidays.
The data scientists are using Amazon Forecast to generate the forecasts.
Which algorithm in Forecast should the data scientists use to meet these requirements?
- A . Autoregressive Integrated Moving Average (AIRMA)
- B . Exponential Smoothing (ETS)
- C . Convolutional Neural Network – Quantile Regression (CNN-QR)
- D . Prophet
Correct Answer: C
C
Explanation:
CNN-QR is a proprietary machine learning algorithm for forecasting time series using causal convolutional neural networks (CNNs). CNN-QR works best with large datasets containing hundreds of time series. It accepts item metadata, and is the only Forecast algorithm that accepts related time series data without future values. In this case, the power company has historical power consumption data for the last 10 years, which is a large dataset with multiple time series. The data also includes related data such as weather, number of individuals on the property, and public holidays, which can be used as item metadata or related time series data. Therefore, CNN-QR is the most suitable algorithm for this scenario.
References: Amazon Forecast Algorithms, Amazon Forecast CNN-QR
C
Explanation:
CNN-QR is a proprietary machine learning algorithm for forecasting time series using causal convolutional neural networks (CNNs). CNN-QR works best with large datasets containing hundreds of time series. It accepts item metadata, and is the only Forecast algorithm that accepts related time series data without future values. In this case, the power company has historical power consumption data for the last 10 years, which is a large dataset with multiple time series. The data also includes related data such as weather, number of individuals on the property, and public holidays, which can be used as item metadata or related time series data. Therefore, CNN-QR is the most suitable algorithm for this scenario.
References: Amazon Forecast Algorithms, Amazon Forecast CNN-QR
Question #298
A company is building a demand forecasting model based on machine learning (ML). In the development stage, an ML specialist uses an Amazon SageMaker notebook to perform feature engineering during work hours that consumes low amounts of CPU and memory resources. A data engineer uses the same notebook to perform data preprocessing once a day on average that requires very high memory and completes in only 2 hours. The data preprocessing is not configured to use GPU. All the processes are running well on an ml.m5.4xlarge notebook instance.
The company receives an AWS Budgets alert that the billing for this month exceeds the allocated budget.
Which solution will result in the MOST cost savings?
- A . Change the notebook instance type to a memory optimized instance with the same vCPU number as the ml.m5.4xlarge instance has. Stop the notebook when it is not in use. Run both data preprocessing and feature engineering development on that instance.
- B . Keep the notebook instance type and size the same. Stop the notebook when it is not in use. Run
data preprocessing on a P3 instance type with the same memory as the ml.m5.4xlarge instance by using Amazon SageMaker Processing. - C . Change the notebook instance type to a smaller general-purpose instance. Stop the notebook when it is not in use. Run data preprocessing on an ml. r5 instance with the same memory size as the ml.m5.4xlarge instance by using Amazon SageMaker Processing.
- D . Change the notebook instance type to a smaller general-purpose instance. Stop the notebook when it is not in use. Run data preprocessing on an R5 instance with the same memory size as the ml.m5.4xlarge instance by using the Reserved Instance option.
Correct Answer: C
C
Explanation:
The best solution to reduce the cost of the notebook instance and the data preprocessing job is to change the notebook instance type to a smaller general-purpose instance, stop the notebook when it is not in use, and run data preprocessing on an ml.r5 instance with the same memory size as the ml.m5.4xlarge instance by using Amazon SageMaker Processing. This solution will result in the most cost savings because:
Changing the notebook instance type to a smaller general-purpose instance will reduce the hourly cost of running the notebook, since the feature engineering development does not require high CPU and memory resources. For example, an ml.t3.medium instance costs $0.0464 per hour, while an ml.m5.4xlarge instance costs $0.888 per hour1.
Stopping the notebook when it is not in use will also reduce the cost, since the notebook will only incur charges when it is running. For example, if the notebook is used for 8 hours per day, 5 days per week, then stopping it when it is not in use will save about 76% of the monthly cost compared to leaving it running all the time2.
Running data preprocessing on an ml.r5 instance with the same memory size as the ml.m5.4xlarge instance by using Amazon SageMaker Processing will reduce the cost of the data preprocessing job, since the ml.r5 instance is optimized for memory-intensive workloads and has a lower cost per GB of memory than the ml.m5 instance. For example, an ml.r5.4xlarge instance has 128 GB of memory and costs $1.008 per hour, while an ml.m5.4xlarge instance has 64 GB of memory and costs $0.888 per hour1. Therefore, the ml.r5.4xlarge instance can process the same amount of data in half the time and at a lower cost than the ml.m5.4xlarge instance. Moreover, using Amazon SageMaker Processing will allow the data preprocessing job to run on a separate, fully managed infrastructure that can be scaled up or down as needed, without affecting the notebook instance. The other options are not as effective as option C for the following reasons:
Option A is not optimal because changing the notebook instance type to a memory optimized instance with the same vCPU number as the ml.m5.4xlarge instance has will not reduce the cost of the notebook, since the memory optimized instances have a higher cost per vCPU than the general-purpose instances. For example, an ml.r5.4xlarge instance has 16 vCPUs and costs $1.008 per hour, while an ml.m5.4xlarge instance has 16 vCPUs and costs $0.888 per hour1. Moreover, running both data preprocessing and feature engineering development on the same instance will not take advantage of the scalability and flexibility of Amazon SageMaker Processing.
Option B is not suitable because running data preprocessing on a P3 instance type with the same memory as the ml.m5.4xlarge instance by using Amazon SageMaker Processing will not reduce the cost of the data preprocessing job, since the P3 instance type is optimized for GPU-based workloads and has a higher cost per GB of memory than the ml.m5 or ml.r5 instance types. For example, an ml.p3.2xlarge instance has 61 GB of memory and costs $3.06 per hour, while an ml.m5.4xlarge instance has 64 GB of memory and costs $0.888 per hour1. Moreover, the data preprocessing job does not require GPU, so using a P3 instance type will be wasteful and inefficient.
Option D is not feasible because running data preprocessing on an R5 instance with the same memory size as the ml.m5.4xlarge instance by using the Reserved Instance option will not reduce the cost of the data preprocessing job, since the Reserved Instance option requires a commitment to a consistent amount of usage for a period of 1 or 3 years3. However, the data preprocessing job only runs once a day on average and completes in only 2 hours, so it does not have a consistent or predictable usage pattern. Therefore, using the Reserved Instance option will not provide any cost savings and may incur additional charges for unused capacity.
References:
Amazon SageMaker Pricing
Manage Notebook Instances – Amazon SageMaker
Amazon EC2 Pricing – Reserved Instances
C
Explanation:
The best solution to reduce the cost of the notebook instance and the data preprocessing job is to change the notebook instance type to a smaller general-purpose instance, stop the notebook when it is not in use, and run data preprocessing on an ml.r5 instance with the same memory size as the ml.m5.4xlarge instance by using Amazon SageMaker Processing. This solution will result in the most cost savings because:
Changing the notebook instance type to a smaller general-purpose instance will reduce the hourly cost of running the notebook, since the feature engineering development does not require high CPU and memory resources. For example, an ml.t3.medium instance costs $0.0464 per hour, while an ml.m5.4xlarge instance costs $0.888 per hour1.
Stopping the notebook when it is not in use will also reduce the cost, since the notebook will only incur charges when it is running. For example, if the notebook is used for 8 hours per day, 5 days per week, then stopping it when it is not in use will save about 76% of the monthly cost compared to leaving it running all the time2.
Running data preprocessing on an ml.r5 instance with the same memory size as the ml.m5.4xlarge instance by using Amazon SageMaker Processing will reduce the cost of the data preprocessing job, since the ml.r5 instance is optimized for memory-intensive workloads and has a lower cost per GB of memory than the ml.m5 instance. For example, an ml.r5.4xlarge instance has 128 GB of memory and costs $1.008 per hour, while an ml.m5.4xlarge instance has 64 GB of memory and costs $0.888 per hour1. Therefore, the ml.r5.4xlarge instance can process the same amount of data in half the time and at a lower cost than the ml.m5.4xlarge instance. Moreover, using Amazon SageMaker Processing will allow the data preprocessing job to run on a separate, fully managed infrastructure that can be scaled up or down as needed, without affecting the notebook instance. The other options are not as effective as option C for the following reasons:
Option A is not optimal because changing the notebook instance type to a memory optimized instance with the same vCPU number as the ml.m5.4xlarge instance has will not reduce the cost of the notebook, since the memory optimized instances have a higher cost per vCPU than the general-purpose instances. For example, an ml.r5.4xlarge instance has 16 vCPUs and costs $1.008 per hour, while an ml.m5.4xlarge instance has 16 vCPUs and costs $0.888 per hour1. Moreover, running both data preprocessing and feature engineering development on the same instance will not take advantage of the scalability and flexibility of Amazon SageMaker Processing.
Option B is not suitable because running data preprocessing on a P3 instance type with the same memory as the ml.m5.4xlarge instance by using Amazon SageMaker Processing will not reduce the cost of the data preprocessing job, since the P3 instance type is optimized for GPU-based workloads and has a higher cost per GB of memory than the ml.m5 or ml.r5 instance types. For example, an ml.p3.2xlarge instance has 61 GB of memory and costs $3.06 per hour, while an ml.m5.4xlarge instance has 64 GB of memory and costs $0.888 per hour1. Moreover, the data preprocessing job does not require GPU, so using a P3 instance type will be wasteful and inefficient.
Option D is not feasible because running data preprocessing on an R5 instance with the same memory size as the ml.m5.4xlarge instance by using the Reserved Instance option will not reduce the cost of the data preprocessing job, since the Reserved Instance option requires a commitment to a consistent amount of usage for a period of 1 or 3 years3. However, the data preprocessing job only runs once a day on average and completes in only 2 hours, so it does not have a consistent or predictable usage pattern. Therefore, using the Reserved Instance option will not provide any cost savings and may incur additional charges for unused capacity.
References:
Amazon SageMaker Pricing
Manage Notebook Instances – Amazon SageMaker
Amazon EC2 Pricing – Reserved Instances
Question #299
The chief editor for a product catalog wants the research and development team to build a machine learning system that can be used to detect whether or not individuals in a collection of images are wearing the company’s retail brand. The team has a set of training data.
Which machine learning algorithm should the researchers use that BEST meets their requirements?
- A . Latent Dirichlet Allocation (LDA)
- B . Recurrent neural network (RNN)
- C . K-means
- D . Convolutional neural network (CNN)
Correct Answer: D
D
Explanation:
The problem of detecting whether or not individuals in a collection of images are wearing the company’s retail brand is an example of image recognition, which is a type of machine learning task that identifies and classifies objects in an image. Convolutional neural networks (CNNs) are a type of machine learning algorithm that are well-suited for image recognition, as they can learn to extract features from images and handle variations in size, shape, color, and orientation of the objects. CNNs consist of multiple layers that perform convolution, pooling, and activation operations on the input images, resulting in a high-level representation that can be used for classification or detection. Therefore, option D is the best choice for the machine learning algorithm that meets the requirements of the chief editor.
Option A is incorrect because latent Dirichlet allocation (LDA) is a type of machine learning algorithm that is used for topic modeling, which is a task that discovers the hidden themes or topics in a collection of text documents. LDA is not suitable for image recognition, as it does not preserve the spatial information of the pixels. Option B is incorrect because recurrent neural networks (RNNs) are a type of machine learning algorithm that are used for sequential data, such as text, speech, or time series. RNNs can learn from the temporal dependencies and patterns in the input data, and generate outputs that depend on the previous states. RNNs are not suitable for image recognition, as they do not capture the spatial dependencies and patterns in the input images. Option C is incorrect because k-means is a type of machine learning algorithm that is used for clustering, which is a task that groups similar data points together based on their features. K-means is not suitable for image recognition, as it does not perform classification or detection of the objects in the images.
References:
Image Recognition Software – ML Image & Video Analysis – Amazon … Image classification and object detection using Amazon Rekognition … AWS Amazon Rekognition – Deep Learning Face and Image Recognition … GitHub – awslabs/aws-ai-solution-kit: Machine Learning APIs for common … Meet iNaturalist, an AWS-powered nature app that helps you identify …
D
Explanation:
The problem of detecting whether or not individuals in a collection of images are wearing the company’s retail brand is an example of image recognition, which is a type of machine learning task that identifies and classifies objects in an image. Convolutional neural networks (CNNs) are a type of machine learning algorithm that are well-suited for image recognition, as they can learn to extract features from images and handle variations in size, shape, color, and orientation of the objects. CNNs consist of multiple layers that perform convolution, pooling, and activation operations on the input images, resulting in a high-level representation that can be used for classification or detection. Therefore, option D is the best choice for the machine learning algorithm that meets the requirements of the chief editor.
Option A is incorrect because latent Dirichlet allocation (LDA) is a type of machine learning algorithm that is used for topic modeling, which is a task that discovers the hidden themes or topics in a collection of text documents. LDA is not suitable for image recognition, as it does not preserve the spatial information of the pixels. Option B is incorrect because recurrent neural networks (RNNs) are a type of machine learning algorithm that are used for sequential data, such as text, speech, or time series. RNNs can learn from the temporal dependencies and patterns in the input data, and generate outputs that depend on the previous states. RNNs are not suitable for image recognition, as they do not capture the spatial dependencies and patterns in the input images. Option C is incorrect because k-means is a type of machine learning algorithm that is used for clustering, which is a task that groups similar data points together based on their features. K-means is not suitable for image recognition, as it does not perform classification or detection of the objects in the images.
References:
Image Recognition Software – ML Image & Video Analysis – Amazon … Image classification and object detection using Amazon Rekognition … AWS Amazon Rekognition – Deep Learning Face and Image Recognition … GitHub – awslabs/aws-ai-solution-kit: Machine Learning APIs for common … Meet iNaturalist, an AWS-powered nature app that helps you identify …
Question #300
An automotive company uses computer vision in its autonomous cars. The company trained its object detection models successfully by using transfer learning from a convolutional neural network (CNN). The company trained the models by using PyTorch through the Amazon SageMaker SDK.
The vehicles have limited hardware and compute power. The company wants to optimize the model to reduce memory, battery, and hardware consumption without a significant sacrifice in accuracy.
Which solution will improve the computational efficiency of the models?
- A . Use Amazon CloudWatch metrics to gain visibility into the SageMaker training weights, gradients, biases, and activation outputs. Compute the filter ranks based on the training information. Apply pruning to remove the low-ranking filters. Set new weights based on the pruned set of filters. Run a new training job with the pruned model.
- B . Use Amazon SageMaker Ground Truth to build and run data labeling workflows. Collect a larger labeled dataset with the labelling workflows. Run a new training job that uses the new labeled data with previous training data.
- C . Use Amazon SageMaker Debugger to gain visibility into the training weights, gradients, biases, and activation outputs. Compute the filter ranks based on the training information. Apply pruning to remove the low-ranking filters. Set the new weights based on the pruned set of filters. Run a new training job with the pruned model.
- D . Use Amazon SageMaker Model Monitor to gain visibility into the ModelLatency metric and OverheadLatency metric of the model after the company deploys the model. Increase the model learning rate. Run a new training job.
Correct Answer: C
C
Explanation:
The solution C will improve the computational efficiency of the models because it uses Amazon SageMaker Debugger and pruning, which are techniques that can reduce the size and complexity of the convolutional neural network (CNN) models.
The solution C involves the following steps:
Use Amazon SageMaker Debugger to gain visibility into the training weights, gradients, biases, and activation outputs. Amazon SageMaker Debugger is a service that can capture and analyze the tensors that are emitted during the training process of machine learning models. Amazon SageMaker Debugger can provide insights into the model performance, quality, and convergence. Amazon SageMaker Debugger can also help to identify and diagnose issues such as overfitting, underfitting, vanishing gradients, and exploding gradients1.
Compute the filter ranks based on the training information. Filter ranking is a technique that can measure the importance of each filter in a convolutional layer based on some criterion, such as the average percentage of zero activations or the L1-norm of the filter weights. Filter ranking can help to identify the filters that have little or no contribution to the model output, and thus can be removed without affecting the model accuracy2.
Apply pruning to remove the low-ranking filters. Pruning is a technique that can reduce the size and complexity of a neural network by removing the redundant or irrelevant parts of the network, such as neurons, connections, or filters. Pruning can help to improve the computational efficiency, memory usage, and inference speed of the model, as well as to prevent overfitting and improve generalization3.
Set the new weights based on the pruned set of filters. After pruning, the model will have a smaller and simpler architecture, with fewer filters in each convolutional layer. The new weights of the model can be set based on the pruned set of filters, either by initializing them randomly or by fine-tuning them from the original weights4.
Run a new training job with the pruned model. The pruned model can be trained again with the same or a different dataset, using the same or a different framework or algorithm. The new training job can use the same or a different configuration of Amazon SageMaker, such as the instance type, the hyperparameters, or the data ingestion mode. The new training job can also use Amazon SageMaker Debugger to monitor and analyze the training process and the model quality5.
The other options are not suitable because:
Option A: Using Amazon CloudWatch metrics to gain visibility into the SageMaker training weights, gradients, biases, and activation outputs will not be as effective as using Amazon SageMaker Debugger. Amazon CloudWatch is a service that can monitor and observe the operational health and performance of AWS resources and applications. Amazon CloudWatch can provide metrics, alarms, dashboards, and logs for various AWS services, including Amazon SageMaker. However, Amazon CloudWatch does not provide the same level of granularity and detail as Amazon SageMaker Debugger for the tensors that are emitted during the training process of machine learning models. Amazon CloudWatch metrics are mainly focused on the resource utilization and the training progress, not on the model performance, quality, and convergence6.
Option B: Using Amazon SageMaker Ground Truth to build and run data labeling workflows and collecting a larger labeled dataset with the labeling workflows will not improve the computational efficiency of the models. Amazon SageMaker Ground Truth is a service that can create high-quality training datasets for machine learning by using human labelers. A larger labeled dataset can help to improve the model accuracy and generalization, but it will not reduce the memory, battery, and hardware consumption of the model. Moreover, a larger labeled dataset may increase the training time and cost of the model7.
Option D: Using Amazon SageMaker Model Monitor to gain visibility into the ModelLatency metric and OverheadLatency metric of the model after the company deploys the model and increasing the model learning rate will not improve the computational efficiency of the models. Amazon SageMaker Model Monitor is a service that can monitor and analyze the quality and performance of machine learning models that are deployed on Amazon SageMaker endpoints. The ModelLatency metric and the OverheadLatency metric can measure the inference latency of the model and the endpoint, respectively. However, these metrics do not provide any information about the training weights, gradients, biases, and activation outputs of the model, which are needed for pruning. Moreover, increasing the model learning rate will not reduce the size and complexity of the model, but it may affect the model convergence and accuracy.
References:
1: Amazon SageMaker Debugger
2: Pruning Convolutional Neural Networks for Resource Efficient Inference
3: Pruning Neural Networks: A Survey
4: Learning both Weights and Connections for Efficient Neural Networks
5: Amazon SageMaker Training Jobs
6: Amazon CloudWatch Metrics for Amazon SageMaker
7: Amazon SageMaker Ground Truth
: Amazon SageMaker Model Monitor
C
Explanation:
The solution C will improve the computational efficiency of the models because it uses Amazon SageMaker Debugger and pruning, which are techniques that can reduce the size and complexity of the convolutional neural network (CNN) models.
The solution C involves the following steps:
Use Amazon SageMaker Debugger to gain visibility into the training weights, gradients, biases, and activation outputs. Amazon SageMaker Debugger is a service that can capture and analyze the tensors that are emitted during the training process of machine learning models. Amazon SageMaker Debugger can provide insights into the model performance, quality, and convergence. Amazon SageMaker Debugger can also help to identify and diagnose issues such as overfitting, underfitting, vanishing gradients, and exploding gradients1.
Compute the filter ranks based on the training information. Filter ranking is a technique that can measure the importance of each filter in a convolutional layer based on some criterion, such as the average percentage of zero activations or the L1-norm of the filter weights. Filter ranking can help to identify the filters that have little or no contribution to the model output, and thus can be removed without affecting the model accuracy2.
Apply pruning to remove the low-ranking filters. Pruning is a technique that can reduce the size and complexity of a neural network by removing the redundant or irrelevant parts of the network, such as neurons, connections, or filters. Pruning can help to improve the computational efficiency, memory usage, and inference speed of the model, as well as to prevent overfitting and improve generalization3.
Set the new weights based on the pruned set of filters. After pruning, the model will have a smaller and simpler architecture, with fewer filters in each convolutional layer. The new weights of the model can be set based on the pruned set of filters, either by initializing them randomly or by fine-tuning them from the original weights4.
Run a new training job with the pruned model. The pruned model can be trained again with the same or a different dataset, using the same or a different framework or algorithm. The new training job can use the same or a different configuration of Amazon SageMaker, such as the instance type, the hyperparameters, or the data ingestion mode. The new training job can also use Amazon SageMaker Debugger to monitor and analyze the training process and the model quality5.
The other options are not suitable because:
Option A: Using Amazon CloudWatch metrics to gain visibility into the SageMaker training weights, gradients, biases, and activation outputs will not be as effective as using Amazon SageMaker Debugger. Amazon CloudWatch is a service that can monitor and observe the operational health and performance of AWS resources and applications. Amazon CloudWatch can provide metrics, alarms, dashboards, and logs for various AWS services, including Amazon SageMaker. However, Amazon CloudWatch does not provide the same level of granularity and detail as Amazon SageMaker Debugger for the tensors that are emitted during the training process of machine learning models. Amazon CloudWatch metrics are mainly focused on the resource utilization and the training progress, not on the model performance, quality, and convergence6.
Option B: Using Amazon SageMaker Ground Truth to build and run data labeling workflows and collecting a larger labeled dataset with the labeling workflows will not improve the computational efficiency of the models. Amazon SageMaker Ground Truth is a service that can create high-quality training datasets for machine learning by using human labelers. A larger labeled dataset can help to improve the model accuracy and generalization, but it will not reduce the memory, battery, and hardware consumption of the model. Moreover, a larger labeled dataset may increase the training time and cost of the model7.
Option D: Using Amazon SageMaker Model Monitor to gain visibility into the ModelLatency metric and OverheadLatency metric of the model after the company deploys the model and increasing the model learning rate will not improve the computational efficiency of the models. Amazon SageMaker Model Monitor is a service that can monitor and analyze the quality and performance of machine learning models that are deployed on Amazon SageMaker endpoints. The ModelLatency metric and the OverheadLatency metric can measure the inference latency of the model and the endpoint, respectively. However, these metrics do not provide any information about the training weights, gradients, biases, and activation outputs of the model, which are needed for pruning. Moreover, increasing the model learning rate will not reduce the size and complexity of the model, but it may affect the model convergence and accuracy.
References:
1: Amazon SageMaker Debugger
2: Pruning Convolutional Neural Networks for Resource Efficient Inference
3: Pruning Neural Networks: A Survey
4: Learning both Weights and Connections for Efficient Neural Networks
5: Amazon SageMaker Training Jobs
6: Amazon CloudWatch Metrics for Amazon SageMaker
7: Amazon SageMaker Ground Truth
: Amazon SageMaker Model Monitor