Stephen

The same format with Adiac will work fine.

1. As the code is using conv2d, so just pad 1 more dimension with 1d kernel. It is the same with (batch_size, sequence_length, sequence_dimension) with conv1d. 2. This is a...

Save the model using tf.keras.model.save function as h5 file or tensorflow graph (recommended), then serving using standard tensorflow protocol. I can add a example later on.