Unknown layer: Dot

[jscriptcoder]

Hi there,

I've seen an issue reported a while ago, #218, where this "Unknown layer: Dot" was mentioned, but there was no a clear solution. Besides the error message seems to be different in my case:

"Uncaught (in promise) Error: Unknown layer: Dot. This may be due to one of the following reasons:

1. The layer is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code.
2. The custom layer is defined in JavaScript, but is not registered properly with tf.serialization.registerClass()."

I'm using TensorFlow.js 0.13.1, in Chrome 68

The keras model trained can be found here: https://github.com/jscriptcoder/date-translator/tree/master/jupyter, dates_model.h5 file, and the converted model is here: https://github.com/jscriptcoder/date-translator/tree/master/tfjsmodel. The Jupyter notebook with the training is here: https://bit.ly/2Oc14Ce.

I'm using Tensorflow 1.10.0 as backend and Keras 2.2.2. To convert the model I used Tensorflowjs 0.6.0.

Now, the error message seems to be quite clear when it says: "The layer is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code", in that case the question is, how can I do so?

Thanks a lot,
Fran

[caisq]

This is a duplicate of #218