Table of Contents
How Google does Machine Learning Week 01 Quiz Answers
Quiz 1: Introduction to ML on Google Cloud
Q1. What is a common reason for an ML model that works well in training but fails in production?
- The model was not properly deployed during production
- Model training was not completed properly
- The wrong model chosen during training
- The ML dataset was improperly created
Q2. Personalized Algorithms are often built using which type of ML model?
- Recommendation systems
- Image classification models
- Sequence models
Q3. What is a key lesson Google has learned with regards to reducing the chance of failure in production ML models?
- Understand and fully utilize TensorFlow
- Base as many models as possible on recommendation systems
- Process batch data and streaming data the same way
Quiz 2: Introduction to AI First
Q1. The main stages of Machine Learning models are?
- Train an ML model
- Predict with a trained model
- Both A and B
- None of the above.
Q2. What are common mathematical models used in Machine Learning?
- Linear methods
- Decision trees
- Radial basis functions
- All the above
Q3. In the past, why did neural networks models have just a few layers?
- Neural networks with lots of layers takes a lot of computing power
- As you add more layers, there are more weights to adjust, and you need lots more data available to make those adjustments.
- If you just add layers, you may run into issues, for example some of the layers may become all zero or blow up and become NAM (not a number).
- All of the above
Q4. What are the models included in Google Translate app?
- Find the sign
- Read the sign
- Detect the language
- All of the above
Q5. What is the smart reply feature of Inbox and Gmail?
- The email program suggests three possible responses to received emails
- The email program automatically writes and sends responses based on past conversations.
- When you hit Send, the email program automatically holds the message and then sends it when the program determines the receiver is most likely to read the message.
- As you respond to a message, the email program suggests words to use based on how smart the receiver is.
Quiz 3: Pre-trained ML APIs
Q1. Which of the following is NOT a pre-trained Machine learning model on Google Cloud?
- Vision API
- Speech API
- Tensorflow
- Translation API
Q2. Which API lets you perform complex image detection with a single REST API request?
- Cloud speech API
- Cloud translation API
- Cloud vision API
- None the above
Q3. Which API lets you understand your video’s entities at shot, frame, or video level?
- Translation API
- Cloud video intelligence API
- Cloud Speech API
- None the above
Q4. What are the benefits of cloud Speech-to-Text API?
- Let’s you perform speech-to-text transcription
- Supports speech timestamps
- Supports profanity filtering
- All of the above
Q5. What type of actions can be done by Cloud Natural Language API?
- Lets you extract entities from your text
- Gives you the overall sentiment of a sentence or a text document
- Gives you the font and heading level used in a sentence or test document
- None of the above
Quiz 4: All about data
Q1. What would you use to replace user input by machine learning?
- Pre-trained models.
- Neural networks.
- Labeled data.
- All of the above.
Q2. Which of the following refers to the type of data used in ML models?
- Labeled data
- Unlabeled data
- Flagged data
- Both A and B
Q3. Which of the following are best practices for Data preparation?
- Avoid training-serving skew
- Avoid target leakage
- Provide a time signal
- All of the above
Q4. Which of the following is not part of the ML training phase?
- Connecting Neural Networks
- Evaluating the models
- Create the models
- Data management
Q5. What’s the most efficient way to transcribe speech?
- You can use a speech API.
- You can collect audio data, train it and predict with it.
- Use a Dictionary website for a partial transcription, then using ML to fill in what’s missing
- All of the above
How Google does Machine Learning Week 02 Quiz Answers
Quiz 1: Transform your business
Q1. Which of the following scenarios may require a supervised learning model to be retrained as a new model?
- The model was trained on unlabeled data and we now wish to train it on labeled data.
- The model was trained on labeled data and we now wish to train it on more to label the data
- The model was trained on unlabeled data and we now wish to add labels to the data.
- The model was trained on labeled data and we now wish to correct the labels of the data.
Q2. A team is preparing to develop and deploy an ML model for use on a shopping website. They have collected a little data to train the model. The team plans on gathering more data once the model is developed. Now they are ready for the next phase, training.
- Which of these scenarios will most likely lead to a successful deployment of the ML model?
- The team should take time to focus on training the perfect model, because deployment is quick and easy.
- The team should take time to gather more data because the quality and architecture of the model are affected by the amount of data
- The team should focus on deployment of the model. The model can be weak to start, then be improved when more user data has been accumulated.
Q3. An online shopping company has a team of customer representatives read emails from customers. Depending upon the content of the email, the representative routes the email to the appropriate department.
- The company would like to alleviate the customer representatives task by automating it. Your team has been asked to create an app to read customer emails and determine which department should handle it.
- Which of these would be a good way to structure the app (choose all that apply)?
- The team should develop one all-encompassing model that will scan the email content, categorize the content, and determine the appropriate team to receive the email.
- The team should develop several models, one for each task. They should develop these models from the ground up and not use pre-existing models, to ensure the models are properly trained.
- Automatic feature extraction
Quiz 2: How Google does ML
Q1. Which of the following networks is used in identifying faces, objects, and traffic signs?
- Convolutional Neural Networks
- Deep Neural Networks
- Recurrent Neural Networks
- None of the above.
Q2. Which of the following statement is true about ML systems?
- It generates a lot of value for the organization, for customers and for end users.
- Almost every single one has a team of people reviewing the algorithms, reviewing their responses and doing random sub-samples.
- Both A and B
- None of the above
Q3. Which of the following are facets that differentiate deep learning networks in multilayer networks?
- More complex ways of connecting layers
- Cambrian explosion of computing power to train
- Automatic feature extraction
- All of the above
Q4. Which of the following statement is incorrect?
- Machine learning performs some core and numerical tasks
- Machine learning doesn’t serve that task in a website.
- Machine learning doesn’t have unit tests of its own.
- None of the above
Quiz 3: Inclusive ML
Q1. Which of the following are correct about Facets?
- It’s an open source data visualization
- Facets was developed at Google and is one of the ways in which you can make machine learning models more inclusive
- Both A and B
- None of the above.
Q2. The things you incorrectly do not predict, things you exclude when instead it should have been included is called?
- False negatives
- False positives
- True positives
- None of the above
Q3. Which of the following help identify areas where a machine learning system could be more inclusive?
- Confusion matrix
- Evaluation metrics
- Both A and B
- None of the above
Q4. Which approach is followed to achieve a better performance across subgroups?
- Equality of opportunity
- Evaluation metrics
- Confusion matrix
- None of the above
Q5. Which of the following are the parts of Facets?
- Overview
- Dive
- Both A and B
- None of the above
Q6. The confusion matrix helps which of the following?
- Understanding inclusion and how to introduce inclusion across different subgroups within your data
- Evaluating performance in machine learning
- Both A and B
- None of the above.
Q7. What do you call the things you incorrectly predicted, and the things you include that aren’t actually in the label and should have instead been excluded?
- False negatives
- False positives
- True positives
- None of the above
How Google does Machine Learning Week 03 Quiz Answers
Week 03: Python Notebooks in the Cloud
Q1. You are going to develop an ML model. You are in Canada and the rest of the team is in Mexico.
Your team wants to use Google Cloud with Python Notebook. Which of the following statements support your decision.
- Cloud AI Platform Notebook runs on virtual machines.
- Cloud AI Platform Notebook is hosted in the cloud.
- Cloud AI Platform Notebook contains both markup and output.
Q2. Your team has decided to use the Compute Engine, Cloud Storage, and Cloud AI Platform Notebook for ML model development
- Which of the following statements are applicable to your situation (choose all that apply)
- You must choose your virtual machine configuration carefully, changing it later will be difficult.
- Every member of the team, regardless of their location, can directly read data fromCloud Storage.
- Latency of data access can be a concern, so carefully select the one for data storage.
Q3. Rewrite this sentence by filling in the blanks with a single word each:
The third wave of cloud is _________________ so you can focus on data ___________ instead of infrastructure.
Word bank: insights, hardware, infrastructure, scalable, cloud-first, serverless, machine learning, Google Cloud, iPython Notebooks
- iPython Notebooks, hardware
- Insights, Google Cloud
- Scalable, hardware
- Serverless, insights
Get all Course Quiz Answers of Machine Learning with TensorFlow on Google Cloud Specialization
How Google does Machine Learning Coursera Quiz Answers
Launching into Machine Learning Coursera Quiz Answers
Introduction to TensorFlow Coursera Quiz Answers
Feature Engineering Coursera Quiz Answers
Art and Science of Machine Learning Coursera Quiz Answers
