A data scientist has explored and sanitized a dataset in preparation for the modeling phase of a supervised learning task. The statistical dispersion can vary widely between features, sometimes by several orders of magnitude. Before moving on to the modeling phase, the data scientist wants to ensure that the prediction performance on the production data is as accurate as possible.
Which sequence of steps should the data scientist take to meet these requirements?

A. Apply random sampling to the dataset. Then split the dataset into training, validation, and test sets.

B. Split the dataset into training, validation, and test sets. Then rescale the training set and apply the same scaling to the validation and test sets.

C. Rescale the dataset. Then split the dataset into training, validation, and test sets.

D. Split the dataset into training, validation, and test sets. Then rescale the training set, the validation set, and the test set independently.

A company is building an application that can predict spam email messages based on email text. The company can generate a few thousand human-labeled datasets that contain a list of email messages and a label of "spam" or "not spam" for each email message. A machine learning (ML) specialist wants to use transfer learning with a Bidirectional Encoder Representations from Transformers (BERT) model that is trained on English Wikipedia text data.
What should the ML specialist do to initialize the model to fine-tune the model with the custom data?

A. Initialize the model with pretrained weights in all layers except the last fully connected layer.

B. Initialize the model with pretrained weights in all layers. Stack a classifier on top of the first output position. Train the classifier with the labeled data.

C. Initialize the model with random weights in all layers. Replace the last fully connected layer with a classifier. Train the classifier with the labeled data.

D. Initialize the model with pretrained weights in all layers. Replace the last fully connected layer with a classifier. Train the classifier with the labeled data.

A Machine Learning Specialist is building a logistic regression model that will predict whether or not a person will order a pizza. The Specialist is trying to build the optimal model with an ideal classification threshold.
What model evaluation technique should the Specialist use to understand how different classification thresholds will impact the model's performance?

A. Receiver operating characteristic (ROC) curve

B. Misclassification rate

C. Root Mean Square Error (RMSE)

D. L1 norm

A company is converting a large number of unstructured paper receipts into images. The company wants to create a model based on natural language processing
(NLP) to find relevant entities such as date, location, and notes, as well as some custom entities such as receipt numbers.
The company is using optical character recognition (OCR) to extract text for data labeling. However, documents are in different structures and formats, and the company is facing challenges with setting up the manual workflows for each document type. Additionally, the company trained a named entity recognition (NER) model for custom entity detection using a small sample size. This model has a very low confidence score and will require retraining with a large dataset.
Which solution for text extraction and entity detection will require the LEAST amount of effort?

A. Extract text from receipt images by using Amazon Textract. Use the Amazon SageMaker BlazingText algorithm to train on the text for entities and custom entities.

B. Extract text from receipt images by using a deep learning OCR model from the AWS Marketplace. Use the NER deep learning model to extract entities.

C. Extract text from receipt images by using Amazon Textract. Use Amazon Comprehend for entity detection, and use Amazon Comprehend custom entity recognition for custom entity detection.

D. Extract text from receipt images by using a deep learning OCR model from the AWS Marketplace. Use Amazon Comprehend for entity detection, and use Amazon Comprehend custom entity recognition for custom entity detection.

A machine learning (ML) engineer has created a feature repository in Amazon SageMaker Feature Store for the company. The company has AWS accounts for development, integration, and production. The company hosts a feature store in the development account. The company uses Amazon S3 buckets to store feature values offline. The company wants to share features and to allow the integration account and the production account to reuse the features that are in the feature repository.
Which combination of steps will meet these requirements? (Choose two.)

A. Create an IAM role in the development account that the integration account and production account can assume. Attach IAM policies to the role that allow access to the feature repository and the S3 buckets.

B. Share the feature repository that is associated the S3 buckets from the development account to the integration account and the production account by using AWS Resource Access Manager (AWS RAM).

C. Use AWS Security Token Service (AWS STS) from the integration account and the production account to retrieve credentials for the development account.

D. Set up S3 replication between the development S3 buckets and the integration and production S3 buckets.

E. Create an AWS PrivateLink endpoint in the development account for SageMaker.

A data scientist is training a text classification model by using the Amazon SageMaker built-in BlazingText algorithm. There are 5 classes in the dataset, with 300 samples for category A, 292 samples for category B, 240 samples for category C, 258 samples for category D, and 310 samples for category E.
The data scientist shuffles the data and splits off 10% for testing. After training the model, the data scientist generates confusion matrices for the training and test sets.
 
 
What could the data scientist conclude form these results?

A. Classes C and D are too similar.

B. The dataset is too small for holdout cross-validation.

C. The data distribution is skewed.

D. The model is overfitting for classes B and E.

A beauty supply store wants to understand some characteristics of visitors to the store. The store has security video recordings from the past several years. The store wants to generate a report of hourly visitors from the recordings. The report should group visitors by hair style and hair color.
Which solution will meet these requirements with the LEAST amount of effort?

A. Use an object detection algorithm to identify a visitor’s hair in video frames. Pass the identified hair to an ResNet-50 algorithm to determine hair style and hair color.

B. Use an object detection algorithm to identify a visitor’s hair in video frames. Pass the identified hair to an XGBoost algorithm to determine hair style and hair color.

C. Use a semantic segmentation algorithm to identify a visitor’s hair in video frames. Pass the identified hair to an ResNet-50 algorithm to determine hair style and hair color.

D. Use a semantic segmentation algorithm to identify a visitor’s hair in video frames. Pass the identified hair to an XGBoost algorithm to determine hair style and hair.

A Machine Learning Specialist is designing a system for improving sales for a company. The objective is to use the large amount of information the company has on users' behavior and product preferences to predict which products users would like based on the users' similarity to other users.
What should the Specialist do to meet this objective?

A. Build a content-based filtering recommendation engine with Apache Spark ML on Amazon EMR

B. Build a collaborative filtering recommendation engine with Apache Spark ML on Amazon EMR.

C. Build a model-based filtering recommendation engine with Apache Spark ML on Amazon EMR

D. Build a combinative filtering recommendation engine with Apache Spark ML on Amazon EMR

A sports broadcasting company is planning to introduce subtitles in multiple languages for a live broadcast. The commentary is in English. The company needs the transcriptions to appear on screen in French or Spanish, depending on the broadcasting country. The transcriptions must be able to capture domain-specific terminology, names, and locations based on the commentary context. The company needs a solution that can support options to provide tuning data.
Which combination of AWS services and features will meet these requirements with the LEAST operational overhead? (Choose two.)

A. Amazon Transcribe with custom vocabularies

B. Amazon Transcribe with custom language models

C. Amazon SageMaker Seq2Seq

D. Amazon SageMaker with Hugging Face Speech2Text

E. Amazon Translate

A car company is developing a machine learning solution to detect whether a car is present in an image. The image dataset consists of one million images. Each image in the dataset is 200 pixels in height by 200 pixels in width. Each image is labeled as either having a car or not having a car.
Which architecture is MOST likely to produce a model that detects whether a car is present in an image with the highest accuracy?

A. Use a deep convolutional neural network (CNN) classifier with the images as input. Include a linear output layer that outputs the probability that an image contains a car.

B. Use a deep convolutional neural network (CNN) classifier with the images as input. Include a softmax output layer that outputs the probability that an image contains a car.

C. Use a deep multilayer perceptron (MLP) classifier with the images as input. Include a linear output layer that outputs the probability that an image contains a car.

D. Use a deep multilayer perceptron (MLP) classifier with the images as input. Include a softmax output layer that outputs the probability that an image contains a car.

A data scientist is evaluating a GluonTS on Amazon SageMaker DeepAR model. The evaluation metrics on the test set indicate that the coverage score is 0.489 and 0.889 at the 0.5 and 0.9 quantiles, respectively.
What can the data scientist reasonably conclude about the distributional forecast related to the test set?

A. The coverage scores indicate that the distributional forecast is poorly calibrated. These scores should be approximately equal to each other at all quantiles.

B. The coverage scores indicate that the distributional forecast is poorly calibrated. These scores should peak at the median and be lower at the tails.

C. The coverage scores indicate that the distributional forecast is correctly calibrated. These scores should always fall below the quantile itself.

D. The coverage scores indicate that the distributional forecast is correctly calibrated. These scores should be approximately equal to the quantile itself.

An aircraft engine manufacturing company is measuring 200 performance metrics in a time-series. Engineers want to detect critical manufacturing defects in near- real time during testing. All of the data needs to be stored for offline analysis.
What approach would be the MOST effective to perform near-real time defect detection?

A. Use AWS IoT Analytics for ingestion, storage, and further analysis. Use Jupyter notebooks from within AWS IoT Analytics to carry out analysis for anomalies.

B. Use Amazon S3 for ingestion, storage, and further analysis. Use an Amazon EMR cluster to carry out Apache Spark ML k-means clustering to determine anomalies.

C. Use Amazon S3 for ingestion, storage, and further analysis. Use the Amazon SageMaker Random Cut Forest (RCF) algorithm to determine anomalies.

D. Use Amazon Kinesis Data Firehose for ingestion and Amazon Kinesis Data Analytics Random Cut Forest (RCF) to perform anomaly detection. Use Kinesis Data Firehose to store data in Amazon S3 for further analysis.

An agricultural company is interested in using machine learning to detect specific types of weeds in a 100-acre grassland field. Currently, the company uses tractor-mounted cameras to capture multiple images of the field as 10 ֳ— 10 grids. The company also has a large training dataset that consists of annotated images of popular weed classes like broadleaf and non-broadleaf docks.
The company wants to build a weed detection model that will detect specific types of weeds and the location of each type within the field. Once the model is ready, it will be hosted on Amazon SageMaker endpoints. The model will perform real-time inferencing using the images captured by the cameras.
Which approach should a Machine Learning Specialist take to obtain accurate predictions?

A. Prepare the images in RecordIO format and upload them to Amazon S3. Use Amazon SageMaker to train, test, and validate the model using an image classification algorithm to categorize images into various weed classes.

B. Prepare the images in Apache Parquet format and upload them to Amazon S3. Use Amazon SageMaker to train, test, and validate the model using an object- detection single-shot multibox detector (SSD) algorithm.

C. Prepare the images in RecordIO format and upload them to Amazon S3. Use Amazon SageMaker to train, test, and validate the model using an object- detection single-shot multibox detector (SSD) algorithm.

D. Prepare the images in Apache Parquet format and upload them to Amazon S3. Use Amazon SageMaker to train, test, and validate the model using an image classification algorithm to categorize images into various weed classes.

A Data Science team within a large company uses Amazon SageMaker notebooks to access data stored in Amazon S3 buckets. The IT Security team is concerned that internet-enabled notebook instances create a security vulnerability where malicious code running on the instances could compromise data privacy.
The company mandates that all instances stay within a secured VPC with no internet access, and data communication traffic must stay within the AWS network.
How should the Data Science team configure the notebook instance placement to meet these requirements?

A. Associate the Amazon SageMaker notebook with a private subnet in a VPC. Place the Amazon SageMaker endpoint and S3 buckets within the same VPC.

B. Associate the Amazon SageMaker notebook with a private subnet in a VPC. Use IAM policies to grant access to Amazon S3 and Amazon SageMaker.

C. Associate the Amazon SageMaker notebook with a private subnet in a VPC. Ensure the VPC has S3 VPC endpoints and Amazon SageMaker VPC endpoints attached to it.

D. Associate the Amazon SageMaker notebook with a private subnet in a VPC. Ensure the VPC has a NAT gateway and an associated security group allowing only outbound connections to Amazon S3 and Amazon SageMaker.

A Data Scientist needs to create a serverless ingestion and analytics solution for high-velocity, real-time streaming data.
The ingestion process must buffer and convert incoming records from JSON to a query-optimized, columnar format without data loss. The output datastore must be highly available, and Analysts must be able to run SQL queries against the data and connect to existing business intelligence dashboards.
Which solution should the Data Scientist build to satisfy the requirements?

A. Create a schema in the AWS Glue Data Catalog of the incoming data format. Use an Amazon Kinesis Data Firehose delivery stream to stream the data and transform the data to Apache Parquet or ORC format using the AWS Glue Data Catalog before delivering to Amazon S3. Have the Analysts query the data directly from Amazon S3 using Amazon Athena, and connect to BI tools using the Athena Java Database Connectivity (JDBC) connector.

B. Write each JSON record to a staging location in Amazon S3. Use the S3 Put event to trigger an AWS Lambda function that transforms the data into Apache Parquet or ORC format and writes the data to a processed data location in Amazon S3. Have the Analysts query the data directly from Amazon S3 using Amazon Athena, and connect to BI tools using the Athena Java Database Connectivity (JDBC) connector.

C. Write each JSON record to a staging location in Amazon S3. Use the S3 Put event to trigger an AWS Lambda function that transforms the data into Apache Parquet or ORC format and inserts it into an Amazon RDS PostgreSQL database. Have the Analysts query and run dashboards from the RDS database.

D. Use Amazon Kinesis Data Analytics to ingest the streaming data and perform real-time SQL queries to convert the records to Apache Parquet before delivering to Amazon S3. Have the Analysts query the data directly from Amazon S3 using Amazon Athena and connect to BI tools using the Athena Java Database Connectivity (JDBC) connector.

An employee found a video clip with audio on a company's social media feed. The language used in the video is Spanish. English is the employee's first language, and they do not understand Spanish. The employee wants to do a sentiment analysis.
What combination of services is the MOST efficient to accomplish the task?

A. Amazon Transcribe, Amazon Translate, and Amazon Comprehend

B. Amazon Transcribe, Amazon Comprehend, and Amazon SageMaker seq2seq

C. Amazon Transcribe, Amazon Translate, and Amazon SageMaker Neural Topic Model (NTM)

D. Amazon Transcribe, Amazon Translate and Amazon SageMaker BlazingText

A social media company wants to develop a machine learning (ML) model to detect inappropriate or offensive content in images. The company has collected a large dataset of labeled images and plans to use the built-in Amazon SageMaker image classification algorithm to train the model. The company also intends to use SageMaker pipe mode to speed up the training.
The company splits the dataset into training, validation, and testing datasets. The company stores the training and validation images in folders that are named Training and Validation, respectively. The folders contain subfolders that correspond to the names of the dataset classes. The company resizes the images to the same size and generates two input manifest files named training.lst and validation.lst, for the training dataset and the validation dataset, respectively. Finally, the company creates two separate Amazon S3 buckets for uploads of the training dataset and the validation dataset.
Which additional data preparation steps should the company take before uploading the files to Amazon S3?

A. Generate two Apache Parquet files, training.parquet and validation.parquet, by reading the images into a Pandas data frame and storing the data frame as a Parquet file. Upload the Parquet files to the training S3 bucket.

B. Compress the training and validation directories by using the Snappy compression library. Upload the manifest and compressed files to the training S3 bucket.

C. Compress the training and validation directories by using the gzip compression library. Upload the manifest and compressed files to the training S3 bucket.

D. Generate two RecordIO files, training.rec and validation.rec, from the manifest files by using the im2rec Apache MXNet utility tool. Upload the RecordIO files to the training S3 bucket.

A credit card company wants to build a credit scoring model to help predict whether a new credit card applicant will default on a credit card payment. The company has collected data from a large number of sources with thousands of raw attributes. Early experiments to train a classification model revealed that many attributes are highly correlated, the large number of features slows down the training speed significantly, and that there are some overfitting issues.
The Data Scientist on this project would like to speed up the model training time without losing a lot of information from the original dataset.
Which feature engineering technique should the Data Scientist use to meet the objectives?

A. Run self-correlation on all features and remove highly correlated features

B. Normalize all numerical values to be between 0 and 1

C. Use an autoencoder or principal component analysis (PCA) to replace original features with new features

D. Cluster raw data using k-means and use sample data from each cluster to build a new dataset

A media company with a very large archive of unlabeled images, text, audio, and video footage wishes to index its assets to allow rapid identification of relevant content by the Research team. The company wants to use machine learning to accelerate the efforts of its in-house researchers who have limited machine learning expertise.
Which is the FASTEST route to index the assets?

A. Use Amazon Rekognition, Amazon Comprehend, and Amazon Transcribe to tag data into distinct categories/classes.

B. Create a set of Amazon Mechanical Turk Human Intelligence Tasks to label all footage.

C. Use Amazon Transcribe to convert speech to text. Use the Amazon SageMaker Neural Topic Model (NTM) and Object Detection algorithms to tag data into distinct categories/classes.

D. Use the AWS Deep Learning AMI and Amazon EC2 GPU instances to create custom models for audio transcription and topic modeling, and use object detection to tag data into distinct categories/classes.

A company's machine learning (ML) specialist is building a computer vision model to classify 10 different traffic signs. The company has stored 100 images of each class in Amazon S3, and the company has another 10,000 unlabeled images. All the images come from dash cameras and are a size of 224 pixels × 224 pixels. After several training runs, the model is overfitting on the training data.
Which actions should the ML specialist take to address this problem? (Choose two.)

A. Use Amazon SageMaker Ground Truth to label the unlabeled images.

B. Use image preprocessing to transform the images into grayscale images.

C. Use data augmentation to rotate and translate the labeled images.

D. Replace the activation of the last layer with a sigmoid.

E. Use the Amazon SageMaker k-nearest neighbors (k-NN) algorithm to label the unlabeled images.

A data scientist must build a custom recommendation model in Amazon SageMaker for an online retail company. Due to the nature of the company's products, customers buy only 4-5 products every 5-10 years. So, the company relies on a steady stream of new customers. When a new customer signs up, the company collects data on the customer's preferences. Below is a sample of the data available to the data scientist.
 
How should the data scientist split the dataset into a training and test set for this use case?

A. Shuffle all interaction data. Split off the last 10% of the interaction data for the test set.

B. Identify the most recent 10% of interactions for each user. Split off these interactions for the test set.

C. Identify the 10% of users with the least interaction data. Split off all interaction data from these users for the test set.

D. Randomly select 10% of the users. Split off all interaction data from these users for the test set.

A company wants to classify user behavior as either fraudulent or normal. Based on internal research, a machine learning specialist will build a binary classifier based on two features: age of account, denoted by x, and transaction month, denoted by y. The class distributions are illustrated in the provided figure. The positive class is portrayed in red, while the negative class is portrayed in black.
 
Which model would have the HIGHEST accuracy?

A. Linear support vector machine (SVM)

B. Decision tree

C. Support vector machine (SVM) with a radial basis function kernel

D. Single perceptron with a Tanh activation function

A Machine Learning Specialist is assigned to a Fraud Detection team and must tune an XGBoost model, which is working appropriately for test data. However, with unknown data, it is not working as expected. The existing parameters are provided as follows.
 
Which parameter tuning guidelines should the Specialist follow to avoid overfitting?

A. Increase the max_depth parameter value.

B. Lower the max_depth parameter value.

C. Update the objective to binary:logistic.

D. Lower the min_child_weight parameter value.

A health care company is planning to use neural networks to classify their X-ray images into normal and abnormal classes. The labeled data is divided into a training set of 1,000 images and a test set of 200 images. The initial training of a neural network model with 50 hidden layers yielded 99% accuracy on the training set, but only 55% accuracy on the test set.
What changes should the Specialist consider to solve this issue? (Choose three.)

A. Choose a higher number of layers

B. Choose a lower number of layers

C. Choose a smaller learning rate

D. Enable dropout

E. Include all the images from the test set in the training set

F. Enable early stopping

A data engineer needs to provide a team of data scientists with the appropriate dataset to run machine learning training jobs. The data will be stored in Amazon S3. The data engineer is obtaining the data from an Amazon Redshift database and is using join queries to extract a single tabular dataset. A portion of the schema is as follows:
TransactionTimestamp (Timestamp)
CardName (Varchar)
CardNo (Varchar)
The data engineer must provide the data so that any row with a CardNo value of NULL is removed. Also, the TransactionTimestamp column must be separated into a TransactionDate column and a TransactionTime column. Finally, the CardName column must be renamed to NameOnCard.
The data will be extracted on a monthly basis and will be loaded into an S3 bucket. The solution must minimize the effort that is needed to set up infrastructure for the ingestion and transformation. The solution also must be automated and must minimize the load on the Amazon Redshift cluster.
Which solution meets these requirements?

A. Set up an Amazon EMR cluster. Create an Apache Spark job to read the data from the Amazon Redshift cluster and transform the data. Load the data into the S3 bucket. Schedule the job to run monthly.

B. Set up an Amazon EC2 instance with a SQL client tool, such as SQL Workbench/J, to query the data from the Amazon Redshift cluster directly Export the resulting dataset into a file. Upload the file into the S3 bucket. Perform these tasks monthly.

C. Set up an AWS Glue job that has the Amazon Redshift cluster as the source and the S3 bucket as the destination. Use the built-in transforms Filter, Map, and RenameField to perform the required transformations. Schedule the job to run monthly.

D. Use Amazon Redshift Spectrum to run a query that writes the data directly to the S3 bucket. Create an AWS Lambda function to run the query monthly.

A data scientist has a dataset of machine part images stored in Amazon Elastic File System (Amazon EFS). The data scientist needs to use Amazon SageMaker to create and train an image classification machine learning model based on this dataset. Because of budget and time constraints, management wants the data scientist to create and train a model with the least number of steps and integration work required.
How should the data scientist meet these requirements?

A. Mount the EFS file system to a SageMaker notebook and run a script that copies the data to an Amazon FSx for Lustre file system. Run the SageMaker training job with the FSx for Lustre file system as the data source.

B. Launch a transient Amazon EMR cluster. Configure steps to mount the EFS file system and copy the data to an Amazon S3 bucket by using S3DistCp. Run the SageMaker training job with Amazon S3 as the data source.

C. Mount the EFS file system to an Amazon EC2 instance and use the AWS CLI to copy the data to an Amazon S3 bucket. Run the SageMaker training job with Amazon S3 as the data source.

D. Run a SageMaker training job with an EFS file system as the data source.

A retail company intends to use machine learning to categorize new products. A labeled dataset of current products was provided to the Data Science team. The dataset includes 1,200 products. The labeled dataset has 15 features for each product such as title dimensions, weight, and price. Each product is labeled as belonging to one of six categories such as books, games, electronics, and movies.
Which model should be used for categorizing new products using the provided dataset for training?

A. AnXGBoost model where the objective parameter is set to multi:softmax

B. A deep convolutional neural network (CNN) with a softmax activation function for the last layer

C. A regression forest where the number of trees is set equal to the number of product categories

D. A DeepAR forecasting model based on a recurrent neural network (RNN)

A machine learning (ML) engineer uses Bayesian optimization for a hyperpara meter tuning job in Amazon SageMaker. The ML engineer uses precision as the objective metric.
The ML engineer wants to use recall as the objective metric. The ML engineer also wants to expand the hyperparameter range for a new hyperparameter tuning job. The new hyperparameter range will include the range of the previously performed tuning job.
Which approach will run the new hyperparameter tuning job in the LEAST amount of time?

A. Use a warm start hyperparameter tuning job.

B. Use a checkpointing hyperparameter tuning job.

C. Use the same random seed for the hyperparameter tuning job.

D. Use multiple jobs in parallel for the hyperparameter tuning job.

A newspaper publisher has a table of customer data that consists of several numerical and categorical features, such as age and education history, as well as subscription status. The company wants to build a targeted marketing model for predicting the subscription status based on the table data.
Which Amazon SageMaker built-in algorithm should be used to model the targeted marketing?

A. Random Cut Forest (RCF)

B. XGBoost

C. Neural Topic Model (NTM)

D. DeepAR forecasting

A retail company uses a machine learning (ML) model for daily sales forecasting. The company's brand manager reports that the model has provided inaccurate results for the past 3 weeks.
At the end of each day, an AWS Glue job consolidates the input data that is used for the forecasting with the actual daily sales data and the predictions of the model. The AWS Glue job stores the data in Amazon S3. The company's ML team is using an Amazon SageMaker Studio notebook to gain an understanding about the source of the model's inaccuracies.
What should the ML team do on the SageMaker Studio notebook to visualize the model's degradation MOST accurately?

A. Create a histogram of the daily sales over the last 3 weeks. In addition, create a histogram of the daily sales from before that period.

B. Create a histogram of the model errors over the last 3 weeks. In addition, create a histogram of the model errors from before that period.

C. Create a line chart with the weekly mean absolute error (MAE) of the model.

D. Create a scatter plot of daily sales versus model error for the last 3 weeks. In addition, create a scatter plot of daily sales versus model error from before that period.

A company is observing low accuracy while training on the default built-in image classification algorithm in Amazon SageMaker. The Data Science team wants to use an Inception neural network architecture instead of a ResNet architecture.
Which of the following will accomplish this? (Choose two.)

A. Customize the built-in image classification algorithm to use Inception and use this for model training.

B. Create a support case with the SageMaker team to change the default image classification algorithm to Inception.

C. Bundle a Docker container with TensorFlow Estimator loaded with an Inception network and use this for model training.

D. Use custom code in Amazon SageMaker with TensorFlow Estimator to load the model with an Inception network, and use this for model training.

E. Download and apt-get install the inception network code into an Amazon EC2 instance and use this instance as a Jupyter notebook in Amazon SageMaker.

A company is using Amazon SageMaker to build a machine learning (ML) model to predict customer churn based on customer call transcripts. Audio files from customer calls are located in an on-premises VoIP system that has petabytes of recorded calls. The on-premises infrastructure has high-velocity networking and connects to the company's AWS infrastructure through a VPN connection over a 100 Mbps connection.
The company has an algorithm for transcribing customer calls that requires GPUs for inference. The company wants to store these transcriptions in an Amazon S3 bucket in the AWS Cloud for model development.
Which solution should an ML specialist use to deliver the transcriptions to the S3 bucket as quickly as possible?

A. Order and use an AWS Snowball Edge Compute Optimized device with an NVIDIA Tesla module to run the transcription algorithm. Use AWS DataSync to send the resulting transcriptions to the transcription S3 bucket.

B. Order and use an AWS Snowcone device with Amazon EC2 Inf1 instances to run the transcription algorithm. Use AWS DataSync to send the resulting transcriptions to the transcription S3 bucket.

C. Order and use AWS Outposts to run the transcription algorithm on GPU-based Amazon EC2 instances. Store the resulting transcriptions in the transcription S3 bucket.

D. Use AWS DataSync to ingest the audio files to Amazon S3. Create an AWS Lambda function to run the transcription algorithm on the audio files when they are uploaded to Amazon S3. Configure the function to write the resulting transcriptions to the transcription S3 bucket.

A Machine Learning Specialist works for a credit card processing company and needs to predict which transactions may be fraudulent in near-real time.
Specifically, the Specialist must train a model that returns the probability that a given transaction may fraudulent.
How should the Specialist frame this business problem?

A. Streaming classification

B. Binary classification

C. Multi-category classification

D. Regression classification

An ecommerce company has developed a XGBoost model in Amazon SageMaker to predict whether a customer will return a purchased item. The dataset is imbalanced. Only 5% of customers return items.
A data scientist must find the hyperparameters to capture as many instances of returned items as possible. The company has a small budget for compute.
How should the data scientist meet these requirements MOST cost-effectively?

A. Tune all possible hyperparameters by using automatic model tuning (AMT). Optimize on {“HyperParameterTuningJobObjective”: {“MetricName”: “validation:accuracy”, “Type”: “Maximize”}}.

B. Tune the csv_weight hyperparameter and the scale_pos_weight hyperparameter by using automatic model tuning (AMT). Optimize on {“HyperParameterTuningJobObjective”: {“MetricName”: “validation’ll”, “Type”: “Maximize”}}.

C. Tune all possible hyperparameters by using automatic model tuning (AMT). Optimize on {“HyperParameterTuningJobObjective”: {“MetricName”: “validation:f1”, “Type”: “Maximize”}}.

D. Tune the csv_weight hyperparameter and the scale_pos_weight hyperparameter by using automatic model tuning (AMT). Optimize on {“HyperParameterTuningJobObjective”: {“MetricName”: “validation:f1”, “Type”: “Minimize”}}.

A library is developing an automatic book-borrowing system that uses Amazon Rekognition. Images of library members' faces are stored in an Amazon S3 bucket.
When members borrow books, the Amazon Rekognition CompareFaces API operation compares real faces against the stored faces in Amazon S3.
The library needs to improve security by making sure that images are encrypted at rest. Also, when the images are used with Amazon Rekognition. they need to be encrypted in transit. The library also must ensure that the images are not used to improve Amazon Rekognition as a service.
How should a machine learning specialist architect the solution to satisfy these requirements?

A. Enable server-side encryption on the S3 bucket. Submit an AWS Support ticket to opt out of allowing images to be used for improving the service, and follow the process provided by AWS Support.

B. Switch to using an Amazon Rekognition collection to store the images. Use the IndexFaces and SearchFacesByImage API operations instead of the CompareFaces API operation.

C. Switch to using the AWS GovCloud (US) Region for Amazon S3 to store images and for Amazon Rekognition to compare faces. Set up a VPN connection and only call the Amazon Rekognition API operations through the VPN.

D. Enable client-side encryption on the S3 bucket. Set up a VPN connection and only call the Amazon Rekognition API operations through the VPN.

An ecommerce company is automating the categorization of its products based on images. A data scientist has trained a computer vision model using the Amazon
SageMaker image classification algorithm. The images for each product are classified according to specific product lines. The accuracy of the model is too low when categorizing new products. All of the product images have the same dimensions and are stored within an Amazon S3 bucket. The company wants to improve the model so it can be used for new products as soon as possible.
Which steps would improve the accuracy of the solution? (Choose three.)

A. Use the SageMaker semantic segmentation algorithm to train a new model to achieve improved accuracy.

B. Use the Amazon Rekognition DetectLabels API to classify the products in the dataset.

C. Augment the images in the dataset. Use open source libraries to crop, resize, flip, rotate, and adjust the brightness and contrast of the images.

D. Use a SageMaker notebook to implement the normalization of pixels and scaling of the images. Store the new dataset in Amazon S3.

E. Use Amazon Rekognition Custom Labels to train a new model.

F. Check whether there are class imbalances in the product categories, and apply oversampling or undersampling as required. Store the new dataset in Amazon S3.

A financial services company is building a robust serverless data lake on Amazon S3. The data lake should be flexible and meet the following requirements:
✑ Support querying old and new data on Amazon S3 through Amazon Athena and Amazon Redshift Spectrum.
✑ Support event-driven ETL pipelines
✑ Provide a quick and easy way to understand metadata
Which approach meets these requirements?

A. Use an AWS Glue crawler to crawl S3 data, an AWS Lambda function to trigger an AWS Glue ETL job, and an AWS Glue Data catalog to search and discover metadata.

B. Use an AWS Glue crawler to crawl S3 data, an AWS Lambda function to trigger an AWS Batch job, and an external Apache Hive metastore to search and discover metadata.

C. Use an AWS Glue crawler to crawl S3 data, an Amazon CloudWatch alarm to trigger an AWS Batch job, and an AWS Glue Data Catalog to search and discover metadata.

D. Use an AWS Glue crawler to crawl S3 data, an Amazon CloudWatch alarm to trigger an AWS Glue ETL job, and an external Apache Hive metastore to search and discover metadata.

A data scientist is building a linear regression model. The scientist inspects the dataset and notices that the mode of the distribution is lower than the median, and the median is lower than the mean.
Which data transformation will give the data scientist the ability to apply a linear regression model?

A. Exponential transformation

B. Logarithmic transformation

C. Polynomial transformation

D. Sinusoidal transformation

A company offers an online shopping service to its customers. The company wants to enhance the site's security by requesting additional information when customers access the site from locations that are different from their normal location. The company wants to update the process to call a machine learning (ML) model to determine when additional information should be requested.
The company has several terabytes of data from its existing ecommerce web servers containing the source IP addresses for each request made to the web server. For authenticated requests, the records also contain the login name of the requesting user.
Which approach should an ML specialist take to implement the new security feature in the web application?

A. Use Amazon SageMaker Ground Truth to label each record as either a successful or failed access attempt. Use Amazon SageMaker to train a binary classification model using the factorization machines (FM) algorithm.

B. Use Amazon SageMaker to train a model using the IP Insights algorithm. Schedule updates and retraining of the model using new log data nightly.

C. Use Amazon SageMaker Ground Truth to label each record as either a successful or failed access attempt. Use Amazon SageMaker to train a binary classification model using the IP Insights algorithm.

D. Use Amazon SageMaker to train a model using the Object2Vec algorithm. Schedule updates and retraining of the model using new log data nightly.

A manufacturer of car engines collects data from cars as they are being driven. The data collected includes timestamp, engine temperature, rotations per minute
(RPM), and other sensor readings. The company wants to predict when an engine is going to have a problem, so it can notify drivers in advance to get engine maintenance. The engine data is loaded into a data lake for training.
Which is the MOST suitable predictive model that can be deployed into production?

A. Add labels over time to indicate which engine faults occur at what time in the future to turn this into a supervised learning problem. Use a recurrent neural network (RNN) to train the model to recognize when an engine might need maintenance for a certain fault.

B. This data requires an unsupervised learning algorithm. Use Amazon SageMaker k-means to cluster the data.

C. Add labels over time to indicate which engine faults occur at what time in the future to turn this into a supervised learning problem. Use a convolutional neural network (CNN) to train the model to recognize when an engine might need maintenance for a certain fault.

D. This data is already formulated as a time series. Use Amazon SageMaker seq2seq to model the time series.

A company sells thousands of products on a public website and wants to automatically identify products with potential durability problems. The company has
1.000 reviews with date, star rating, review text, review summary, and customer email fields, but many reviews are incomplete and have empty fields. Each review has already been labeled with the correct durability result.
A machine learning specialist must train a model to identify reviews expressing concerns over product durability. The first model needs to be trained and ready to review in 2 days.
What is the MOST direct approach to solve this problem within 2 days?

A. Train a custom classifier by using Amazon Comprehend.

B. Build a recurrent neural network (RNN) in Amazon SageMaker by using Gluon and Apache MXNet.

C. Train a built-in BlazingText model using Word2Vec mode in Amazon SageMaker.

D. Use a built-in seq2seq model in Amazon SageMaker.

A trucking company is collecting live image data from its fleet of trucks across the globe. The data is growing rapidly and approximately 100 GB of new data is generated every day. The company wants to explore machine learning uses cases while ensuring the data is only accessible to specific IAM users.
Which storage option provides the most processing flexibility and will allow access control with IAM?

A. Use a database, such as Amazon DynamoDB, to store the images, and set the IAM policies to restrict access to only the desired IAM users.

B. Use an Amazon S3-backed data lake to store the raw images, and set up the permissions using bucket policies.

C. Setup up Amazon EMR with Hadoop Distributed File System (HDFS) to store the files, and restrict access to the EMR instances using IAM policies.

D. Configure Amazon EFS with IAM policies to make the data available to Amazon EC2 instances owned by the IAM users.

A Data Science team is designing a dataset repository where it will store a large amount of training data commonly used in its machine learning models. As Data
Scientists may create an arbitrary number of new datasets every day, the solution has to scale automatically and be cost-effective. Also, it must be possible to explore the data using SQL.
Which storage scheme is MOST adapted to this scenario?

A. Store datasets as files in Amazon S3.

B. Store datasets as files in an Amazon EBS volume attached to an Amazon EC2 instance.

C. Store datasets as tables in a multi-node Amazon Redshift cluster.

D. Store datasets as global tables in Amazon DynamoDB.

An ecommerce company is collecting structured data and unstructured data from its website, mobile apps, and IoT devices. The data is stored in several databases and Amazon S3 buckets. The company is implementing a scalable repository to store structured data and unstructured data. The company must implement a solution that provides a central data catalog, self-service access to the data, and granular data access policies and encryption to protect the data.
Which combination of actions will meet these requirements with the LEAST amount of setup? (Choose three.)

A. Identify the existing data in the databases and S3 buckets. Link the data to AWS Lake Formation.

B. Identify the existing data in the databases and S3 buckets. Link the data to AWS Glue.

C. Run AWS Glue crawlers on the linked data sources to create a central data catalog.

D. Apply granular access policies by using AWS Identity and Access Management (1AM). Configure server-side encryption on each data source.

E. Apply granular access policies and encryption by using AWS Lake Formation.

F. Apply granular access policies and encryption by using AWS Glue.

A Machine Learning Specialist previously trained a logistic regression model using scikit-learn on a local machine, and the Specialist now wants to deploy it to production for inference only.
What steps should be taken to ensure Amazon SageMaker can host a model that was trained locally?

A. Build the Docker image with the inference code. Tag the Docker image with the registry hostname and upload it to Amazon ECR.

B. Serialize the trained model so the format is compressed for deployment. Tag the Docker image with the registry hostname and upload it to Amazon S3.

C. Serialize the trained model so the format is compressed for deployment. Build the image and upload it to Docker Hub.

D. Build the Docker image with the inference code. Configure Docker Hub and upload the image to Amazon ECR.

A large consumer goods manufacturer has the following products on sale:
* 34 different toothpaste variants
* 48 different toothbrush variants
* 43 different mouthwash variants
The entire sales history of all these products is available in Amazon S3. Currently, the company is using custom-built autoregressive integrated moving average
(ARIMA) models to forecast demand for these products. The company wants to predict the demand for a new product that will soon be launched.
Which solution should a Machine Learning Specialist apply?

A. Train a custom ARIMA model to forecast demand for the new product.

B. Train an Amazon SageMaker DeepAR algorithm to forecast demand for the new product.

C. Train an Amazon SageMaker k-means clustering algorithm to forecast demand for the new product.

D. Train a custom XGBoost model to forecast demand for the new product.

A data scientist is using Amazon Comprehend to perform sentiment analysis on a dataset of one million social media posts.
Which approach will process the dataset in the LEAST time?

A. Use a combination of AWS Step Functions and an AWS Lambda function to call the DetectSentiment API operation for each post synchronously.

B. Use a combination of AWS Step Functions and an AWS Lambda function to call the BatchDetectSentiment API operation with batches of up to 25 posts at a time.

C. Upload the posts to Amazon S3. Pass the S3 storage path to an AWS Lambda function that calls the StartSentimentDetectionJob API operation.

D. Use an AWS Lambda function to call the BatchDetectSentiment API operation with the whole dataset.

A Data Scientist is training a multilayer perception (MLP) on a dataset with multiple classes. The target class of interest is unique compared to the other classes within the dataset, but it does not achieve and acceptable recall metric. The Data Scientist has already tried varying the number and size of the MLP's hidden layers, which has not significantly improved the results. A solution to improve recall must be implemented as quickly as possible.
Which techniques should be used to meet these requirements?

A. Gather more data using Amazon Mechanical Turk and then retrain

B. Train an anomaly detection model instead of an MLP

C. Train an XGBoost model instead of an MLP

D. Add class weights to the MLP’s loss function and then retrain

A company operates large cranes at a busy port The company plans to use machine learning (ML) for predictive maintenance of the cranes to avoid unexpected breakdowns and to improve productivity.
The company already uses sensor data from each crane to monitor the health of the cranes in real time. The sensor data includes rotation speed, tension, energy consumption, vibration, pressure, and temperature for each crane. The company contracts AWS ML experts to implement an ML solution.
Which potential findings would indicate that an ML-based solution is suitable for this scenario? (Choose two.)

A. The historical sensor data does not include a significant number of data points and attributes for certain time periods.

B. The historical sensor data shows that simple rule-based thresholds can predict crane failures.

C. The historical sensor data contains failure data for only one type of crane model that is in operation and lacks failure data of most other types of crane that are in operation.

D. The historical sensor data from the cranes are available with high granularity for the last 3 years.

E. The historical sensor data contains most common types of crane failures that the company wants to predict.

A data scientist for a medical diagnostic testing company has developed a machine learning (ML) model to identify patients who have a specific disease. The dataset that the scientist used to train the model is imbalanced. The dataset contains a large number of healthy patients and only a small number of patients who have the disease. The model should consider that patients who are incorrectly identified as positive for the disease will increase costs for the company.
Which metric will MOST accurately evaluate the performance of this model?

A. Recall

B. F1 score

C. Accuracy

D. Precision