Overview
- Google BigQuery ML is the company’s cloud offering that will help data analysts (and data scientists) build models through SQL
- BigQuery ML eliminates the need to move your data from the data warehouse, hence speeding up the model building process
- Google has even released guides on how to get started with this, for both data analysts and data scientists
Introduction
Machine learning is THE buzzword of this decade. But job vacancies for experts in this field keeps increasing despite a gigantic amount of people wanting to get a piece of the pie. This is because machine learning is a complex field that combines knowledge, skill, innovation and domain expertise – and most aspiring data scientists are found wanting in at least one of these areas.
Source: Twitter
Most of these aspiring data scientists are coming from an IT background, with a good (or at least manageable) grasp of SQL. If you are one of these people, there’s good news – Google wants to help you get familiar with ML! The tech giant has released Google BigQuery ML, a cloud offering that enables developers to design and build models in BigQuery using standard SQL queries.
Check out the below GIF, posted on Google’s AI blog today, that gives a quick overview of how you can use the tool:
Below are some of the advantages this tool currently offers:
- You don’t need to use Python to build models; SQL is the language BigQuery knows and uses to build and access models
- Data engineers and data analysts who use SQL on a frequent basis can use this with their pre-existing BI tools and spreadsheets
- Of course this helps speed up the model building process significantly as you no longer need to extract and export your data from a data warehouse!
Google has even documented a ‘getting started’ guide for both data analysts and data scientists who want to learn about this tool:
Our take on this
Awesome news for SQL users! I personally know a lot of folks in IT who want to make the transition to machine learning, but are held back by the complexities this field has. BigQuery ML will clear at least a few of those hurdles for you. Learning and doing ML in a language you are absolutely familiar with is priceless.
Currently, BigQuery ML supports both linear regression and logistic regression models but they are working to integrate other techniques into the tool. We should expect to see that happen in the near future.
Subscribe to AVBytes here to get regular data science, machine learning and AI updates in your inbox!
Senior Editor at Analytics Vidhya.Data visualization practitioner who loves reading and delving deeper into the data science and machine learning arts. Always looking for new ways to improve processes using ML and AI.
