Google is learning the cooking machine in his BigQuery data warehouse – TechCrunch

There are still many obstacles to building machine learning models and one of them is that to build these models, developers often have to move a lot of data between their data warehouses and wherever they build their models. Google is now facilitating this part of the process for developers and data specialists of its ecosystem with BigQuery ML, a new feature of its BigQuery data warehouse, by integrating some automatic learning features directly into BigQuery.

With the help of BigQuery ML, developers can create models using linear and logistic regression directly into their data warehouse without having to transfer data backwards while they build and refine their data. models. And all they have to do to build these models and get predictions, is to write a little bit of SQL.

Moving data does not seem like a big deal, but developers often spend a lot of time on this kind of grunt work – time that would be better spent actually working on their models.

BigQuery ML also promises to make building these models easier, even for developers who do not have a lot of experience with machine learning. For starters, developers can use what is basically a standard SQL variant to tell what kind of model they are trying to build and what input data is supposed to be. From there, BigQuery ML then builds the model and allows developers to generate predictions based on it almost immediately. And they will not even have to write code in R or Python.

These new features are now available in beta.