RSquare TypeError: tf__update_state() got an unexpected keyword argument 'sample_weight'

[lotest]

System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): ubuntu 18.04
• TensorFlow version and how it was installed (source or binary): 2.1.0 binary (conda)
• TensorFlow-Addons version and how it was installed (source or binary): 0.8.3 binary(pip)
• Python version: 3.7.6
• Is GPU used? (yes/no): yes

Describe the bug

The code goes wrong when I add "tfa.metrics.RSquare(dtype=tf.float32)" to model metrics.
The exception is "TypeError: tf__update_state() got an unexpected keyword argument 'sample_weight'"

And I don't see "sample_weight" parameter , what been added shown in #564 , of update_state() function in class RSquare in addons version 0.8.3.

Are there something wrong with my installed tensorflow and addons package?

Code to reproduce the issue

Usage in my code:

model.compile(
        loss='mse',
        optimizer=optimizer,
        metrics=['mae', 'mse', tfa.metrics.RSquare(dtype=tf.float32)]
    )

Other info / logs

Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.

[gabrieldemarmiesse]

@lotest #564 was not merged, that's why you don't see sample_weight being available. The problem here is that some of the metrics in TFA do not take a sample_weight argument because they don't use it. It's a problem though because Keras will automatically pass sample weights when the user declare them. A good fix would be to add sample_weights in all metrics and enforce that the argument is present. And if the tensor is not used during the computation, then just raise a warning.

This is linked to #992 , we lack some unified tests. This will become easier once we switch to pytest for testing.

[SSaishruthi]

@gabrieldemarmiesse will take a look at it