mixed precision

[caishiqing]

It seems that the package dose not support mixed precision training, since the Embedding layer needs add operating with output and embedding parameters, which dtype are 'float32' and 'float16' leading to a dtype error "TypeError: Input 'y' of 'AddV2' Op has type float16 that does not match type float32 of argument 'x'.".
Is there any plane to fix this?

[jeisinge]

It is interesting to note that we were able to use mixed-precision training before with TF 1.15 and TF Estimator. With TF 2.3 and Keras, we are seeing this same error.

[callzhang]

Any updates on this?

[leviswind]

Any updates on this?