Get Getting Started with AWS Machine Learning Coursera Quiz Answers
Getting Started with AWS Machine Learning Coursera Quiz Answers
Week 1: Getting Started with AWS Machine Learning
Q1. What is relationship between Artificial intelligence (AI), machine learning (ML) and deep learning (DL)?
- There is no relationship betweeen AI, ML and DL
- AI > ML > DL
- DL > AI > ML
ML > AI > DL
Q2. A company is using a rules based system to classify the credit card transations as fraud or not fraud. Do you think this a machine learning solution?
Q3. How do you define the term artificial intelligence, machine learning and deep learning in one sentence each?
What do you think?
Q4. A security and transportation analysis company wants to use their client’s internal employee images to classify whether to let the employees through the security gate into their client offices. What machine learning algorithm should they use to create this machine learning model?
- Long Short Term Memory Network (LSTM)
- Recurrent Neuran Network (RNN)
- Convolutional Neural Network (CNN)
- Feed-forward Neural Network
Q5. What were the main factors in massive adoption of deep learning in the recent decade? Select two answers.
- Faster and more efficient compute (GPUs)
- More data
- Public Cloud offerings
- Machine learning education and number of data scientists
Week 2: Getting Started with AWS Machine Learning
Q1. In the following confusion matrix for a fredit card detection problem, what is the recall score for the machine learning model?
|header||True Positive||True Negative|
Q2. In general for bias variance tradeoff, the total accuracy is equal to the sum of variance and the square of bias. In the following diagram, pick which line denotes the total accuracy, bias^2 and variance?
- Green line: Bias^2, Orange line: Variance and black line: total error
- Green line: Variance, Orange line: Bias^2 and black line: Variance
- Green line: Bias^2, Orange line: total error and black line: Variance
- Green line: total error, Orange line: Variance and black line: Bias^2
Week 3: Getting Started with AWS Machine Learning
Q1. You need to have deep learning expertise to create image metadata and recognize faces using Amazon Rekognition?
Q2. Which of the following is NOT a good use case for using Amazon Rekogition?
- Object Detection
- Scene Detection
- Fraud Detection
- Text in image detection
Q3. A company wants to use machine learning to detect the incoming text in images to create a searchable video library. They want to extract metadata from the images and index the metadata that can be searched by users easily with minimum management of the proposed pipeline. What should you as an architect recommend their solution be?
- Use Amazon SageMaker to create a CNN model to detect text in an image and use Amazon DynamoDB to index the files
- Use Amazon EMR to create a CNN model to detect text in an image and use Amazon DynamoDB to index the files
- Use Amazon Rekognition to detect text in an images and use Amazon Elasticsearch Service to index the files
- Use Amazon Rekognition to to detect text in an images and use the DetectedText() API to search the index
Week 4: Getting Started with AWS Machine Learning
Q1. A airline booking company wants to determine the sentiment of customers using the support tickets to classify whether the issue is urgent or not and in the process rate the service agents based on the customer feedback. What service can you use to create this solution?
- Amazon Rekognition
- Amazon Comprehend
- Amazon Textract
- Amazon Personalize
Q2. A news organization wants to have transcription of their prime time news for video on-demand service. They are looking for a solution that provides transcription of their news in various languages with minimum management. What AWS services can they use to to create this solution?
- Use Amazon Transcribe to transcribe and Amazon Translate to translate the videos
- Use Amazon Transcribe to transcribe and Amazon Polly translate the videos
- Use Amazon Translate to transcribe and translate the videos
- Use Amazon Lex to transcribe and translate the videos
Week 5: Getting Started with AWS Machine Learning
Q1. Amazon SageMaker can be used in following stages of machine learning pipeline? Select all that apply.
- Model evaluation
- Problem formulation
- Feature engineeringFeature engineering
- Model training
Q2. What is the best way to describe the service Amazon SageMaker?
- Amazon SageMaker is an API blackbox service that helps customers create their machine learning algorithms
- Amazon SageMaker is a fully managed platform service that helps customers through each stage of their machine learning pipeline from creating datasets, training, optimiznig and deploying their machine
- Amazon SageMaker is a framework that helps customers create neural network algorithms and deploy the models
- Amazon SageMaker is a service that creates, test and validates and deploys your models using Pytorch
Q3. What does Amazon SageMaker provide to help developers build their machine learning models and pipeline? Select all that apply.
- Amazon SageMaker provides notebook instances with preinstalled frameworks of to help developers during development
- Amazon SageMaker provides training instances for the creation and training of machine learning models
- Amazon SageMaker provides developers the ability to deploy their models to an endpoint
- Amazon SageMaker provides AutoML for developers to create models using on click
Q1. In which order would you use the different components of Amazon SageMaker in a machine learning pipeline?
- Amazon SageMaker Neo> Amazon SageMaker Training Jobs> Amazon SageMaker Hyperparameter Tuning Jobs > Amazon SageMaker Ground Truth
- Amazon SageMaker Training Jobs > Amazon SageMaker Ground Truth> Amazon SageMaker Hyperparameter Tuning Jobs > Amazon SageMaker Neo
- Amazon Sage Maker Ground Truth> Amazon SageMaker Training Jobs> Amazon SageMaker Hyperparameter Tuning Jobs > Amazon SageMaker Neo
- Amazon SageMaker Neo> Amazon SageMaker Hyperparameter Tuning Jobs> Amazon SageMaker HTraining Job> Amazon SageMaker GroundTruth
Q2. A oil and natural gas organization trained their CNN models using Amazon SageMaker python SDK. What function should they use to create an automatic hyperparameter tuning job in Amazon SageMaker SDK?
Q1. For text classification tasks, what is the best storage layer to use along with Amazon SageMaker?
- Amazon S3
- Amazon Glue
- Amazon EFS
- Amazon EBS
Q2. Amazon BlazingText is an implemention of which algorithm?