BigQuery ML: Machine Learning with Standard SQL

In this episode of AI Adventures, Yufeng introduces BigQuery ML, which allows you to build machine learning models right within BigQuery, using SQL! That means no more copying big tables of data around, and more chances to find insights to your dataset!
BigQuery ML → goo.gle/2m03F5r
Watch the BigQuery Public Datasets episode → goo.gle/2lTRtTY
Check out the rest of the Cloud AI Adventures playlist → goo.gl/UC5usG
Subscribe to get all the episodes as they come out → goo.gle/GCP
product:BigQuery; full name:Yufeng Guo;
#AIAdventures

[@joseortiz_io]

This is crazy! I didn't know about this. Love it. You guys are killing it. I gotta try this out soon on my new RUclips channel! I wish everyone a splendid day! 😁👍

[@gav240z]

I worked with GA for years and have just started on BigQuery, but this is a good example of how useful this can be.

[@samkazery7517]

Guo... you're great at explaining and teaching. Please keep these coming - thank you!

[@RobinYoulton]

Thank you Yufeng, excellent presentation, BigQuery ML is certainly a fantastic tool.

[@jwhite1202]

This is a good starter example to help understand BQ ML for Linear Regression. I plan to work through the other examples too. Good job Google 👍🏾.

[@economyxai]

Makes life easier to have this where the data is, that's for sure. Great idea!

[@pjerrots]

Wow - how brilliant is that!?! I had no idea you could do this! Just next month I have been assigned to building a churn model for a client whose data resides in BQ. If I can get a good model from the logistic regression, it will make everything so much easier! Thanks! :)

[@prettypinksg]

Superb !

[@shanthababu]

Awesome!

[@antoniomariano3797]

typo at 0:38, should be linear regression

[@DannyWaites]

4 years later its still there

[@arjunk5959]

Did not knew anything about BQML before.. Now getting an idea

[@will16320]

It looks like there are a few mistakes in this video, the date range is not updated in the query for instance on ML.evaluate

[@adityachoudhury293]

Also the model names are altered, prediction date ranges need to change too.

[@ayushthakur736]

There might be an error at 0.43s. Shouldn't that be linear regression and not logistic regression. 😊

[@mikemagss]

Slide mistake at 0:46 the first item should read 'Linear Regression'

[@ioia]

Haha, 1:10 may be a typo for the first line. * Liner instead of Logistics

[@yawarkhalid]

Could you maybe link the public dataset that you are using here?

[@TimeoutMegagameplays]

How do you get your data from Cloud SQL to be used in BigQuery for ML?

[@Eng.Omaima]

please if you get the answers share it with me... please I need it in my internship

[@frankfuxa6383]

Do you need a GA360 licence to query this kind of combined user data?

[@gav240z]

Yes, 360 version of GA offers export to BQ functionality. Although App+Web properties may support it similar to Firebase.

[@shaikan0]

You could also use something like Stitchdata to import the data from Google Analytics into GCP as BigQuery tables. It's a lot messier than the native integration with GA360 but you could also perform this type of analysis at a much lower cost.

[@jasonghm]

My question is, Can I predicted sales in the future with this model?

[@franciscaveronicavera2029]

Eu não sei falar lngles