Distributed Processing and Components (TensorFlow Extended)

[@ar-ienterprise3011]

I love your calmness

[@mikeisaacs335]

Excellent episodes giving a good picture into understanding TFX ecosystem

[@anando304]

Excellent video - very insightful to aid ontop of the tfx homepage website

[@azihna]

are you planning on sharing a fully working example of all of the components described in the series?

[@robertcrowe7000]

Yes! Check out the Developer Tutorial for a full working example: www.tensorflow.org/tfx/tutorials/tfx/workshop

[@AbhishekKalgutkar]

Has TFX added support for 2.x models?

[@PavanKumar-rw3br]

Excellent episode... Is there a tutorial for running TFX on premise that doesn't use TensorFlow? Any git code examples, books will greatly help All the TFX examples use Tensorflow. Is TensorFlow pre-requisite for running TFX? I am using pytorch

[@siddharthkotwal7149]

How can ExampleValidator generator an anomaly report without having access to the data? The only two inputs are Schema + Statistics from the previous two components.

[@robertcrowe7000]

The statistics provide sufficient information about the data in order to detect whether the schema constraints are violated (any such violations are what we call anomalies). Note that TFDV (the underlying library) has the ability to return samples of the data corresponding to different anomalies, but this functionality is not currently integrated with TFX.

[@reacionario2026]

Is there one or more books on TFX? And what about running TFX on premises? Or only at google cloud?

[@robertcrowe7000]

No books yet. You can run TFX on premises now, as well as on the Google Cloud and/or Kubeflow.

[@brycejianchen2795]

Great lectures, thank you !

[@gusttavo_kaczynki]

good!

[@alessiasacchi5454]

Crystal clear

[@purushyb]

Liked it