TFLiteToFirebaseML Component

[deep-diver]

Firebase ML is a great place to store TFLite model.

With Firebase ML, we can guarantee that mobile devices can be equipped with the latest ML model without explicitly embedding binary in project compiling stage. We can even A/B test different versions of a model with Google Analytics.

It would be nice if there is a TFX component that helps publishing TFLite model to Firebase ML.

What do you think?

[deep-diver]

@rcrowe-google

so this is equivalent to FirebasePublisher mentioned in tensorflow/tfx#4125

[rcrowe-google]

It sounds like a great idea to me! Have you joined https://groups.google.com/a/tensorflow.org/g/sig-tfx-addons ?

[deep-diver]

@rcrowe-google

yes! I have joined the group already :)
What are the next steps to proceed?

[sayakpaul]

For anyone that might be wondering how well TFLite and Firebase go together here's a blog post:

https://blog.tensorflow.org/2020/06/enhance-your-tensorflow-lite-deployment-with-firebase.html

[rcrowe-google]

@rcrowe-google

yes! I have joined the group already :)
What are the next steps to proceed?

You need to write a proposal. See the proposals README