18 Jan
2022
18 Jan
'22
1:46 p.m.
forward function of perceivermodel, when conditioned with only input_preprocessor and decoder, pseudocode: # step 1 inputs and dimension info = input_preprocessor(inputs) # step 2 encoder_outputs = self.encoder( self.embeddings(batch_size) inputs ) # step 3 decoder_outputs = decoder( decoder.decoder_query(inputs), z=encoder_outputs[0] ) # logits would be additionally processed through output_postprocessor if one were provided return decoder_outputs.logits