Hi, 'm looking at prior and I’m interested in the possibility to condition the generation using external inputs as shown in this image
To do so the best option seems to use a prior with an econder_decoder_continuous configuration, but I’m reading that The configurations decoder_only and recurrent are readily usable, the seq2seq variants depends on another project called rave2vec that will be open sourced in the near future.
Does anybody know if the option to train with the encoder_decoder_continuous configuration is available or is it coming any time soon?
Thanks