seq2seq

source code

data_utils.sentence_to_token_ids
Convert a string to list of integers representing token-ids.

Seq2SeqModel.get_batch(self, data, bucket_id)

• Get a random batch of data from the specified bucket, prepare for step.
• Returns: (encoder_inputs, decoder_inputs, target_weights)
  • encoder_inputs - (bucket_id_encoder_size * batch_size)
  • decoder_inputs - (bucket_id_decoder_size * batch_size)
  • target_weights - (bucket_id_decoder_size * batch_size), create target_weights to be 0 for targets that are padding.