Machine Learning Options on Google Cloud

Question 1

What are options for building machine learning models available on Google Cloud?

Answer 1

• BigQueryML (using SQL queries to create and execute ML models in BigQuery)
• use pre-built APIs (leverage ML models that have already been built and trained by Google)
• AutoML (a no-code solution, so you can build your own ML models on Vertex AI through a point-and-click interface)
• custom training (code your very own ML environment, the training, and the deployment)

Question 2

Which GCP ML option only supports tabular data?

Answer 2

BigQuery ML

Question 3

Which GCP ML option does not require any training data?

Answer 3

pre-built APIs

Question 4

Which GCP ML option does not require Machine learning and coding expertise?

Answer 4

Pre-Built APIs and AutoML

Question 5

What should be used to build custom models with your own training data while spending minimal time coding?

Answer 5

AutoML

Question 6

What should be used if your ML engineers and data scientists want full control of ML workflow?

Answer 6

Vertex AI custom training lets you train and serve custom models with code on Vertex Workbench.

Question 7

Give a short list of GCP pre-built APIs

Answer 7

• Speech-to-Text API (converts audio to text for data processing)
• Cloud Natural Language API (recognizes parts of speech called entities and sentiment)
• Cloud Translation API (converts text from one language to another)
• Text-to-Speech API (converts text into high quality voice audio)
• Vision API (works with and recognizes content in static images)
• Video Intelligence API (recognizes motion and action in video)

Question 8

Describe Auto ML technologies.

Answer 8

1) transfer learning (people with smaller datasets, or less computational power, achieve state-of-the-art results by taking advantage of pre-trained models that have been trained on similar, larger data sets.)
2) neural architecture search (finding the optimal model for the relevant project)

Question 9

List the advantages of AutoML.

Answer 9

• powered by the latest machine-learning research
• trains and evaluates multiple models and compares them to each other
• produces an ensemble of ML models
• chooses the best one
• no code solution

Question 10

What kind of data is supported in AutoML?

Answer 10

AutoML supports four types of data: image, tabular, text, and video

Question 11

What is the major difference between pre-built APIs and AutoML?

Answer 11

The major difference is that pre-built APIs use pre-built machine learning models, while AutoML uses custom-built models

Question 12

What should be used to code custom machine learning model?

Answer 12

Vertex AI Workbench

Question 13

What is Vertex AI Workbench?

Answer 13

Workbench is a single development environment for the entire data science workflow,

Question 14

What is Vertex AI?

Answer 14

a unified platform
It means having one digital experience to create, deploy, and manage models over time, and at scale.

Vertex AI allows users to build machine learning models with either AutoML, a code-less solution or Custom Training,