The Path From Cloud AutoML to Custom Model (Cloud Next '19)

What comes after AutoML? You've created some models in Cloud AutoML, and they've been useful. But you want to see if there's some room for more improvement and customization. Let's see how to start building custom models and deploy them in production. You'll learn how to take advantage of state-of-the-art models, all while advancing your understanding of your data pipelines and machine learning.

Watch more:

Speaker(s): Yufeng Guo, Sara Robinson

Session ID: MLAI301

event: Google Cloud Next 2019; re_ty: Publish; product: TensorFlow - General, Cloud - Data Analytics - BigQuery, Cloud - AI and Machine Learning - Cloud Natural Language, Cloud - AI and Machine Learning - AutoML; fullname: Sara Robinson; event: Google Cloud Next 2019;

[nextwave]

Thanks Google, you are really making whole world better!

[kelvinksau]

Is the code sample available to public ?

[OscarFraxedas]

Did you compare the autoML solution VS the "bag of words" implementation?
It will be interesting to know how good autoML is out of the box.

[zeshan]

would it be possible for further interpreting model trained by autoML?

[kasingau]

Is the sample Colab notebook available to public ?

[SiddheshBhurke]

ml-engine is no longer available. Could you update the code with ai-platform?

[louiswang]

may i get the link for the colab notebook? Thanks!