Decoder输出部分的代码

[shaneshbi]

inference_decoder_output, _, _ = tf.contrib.seq2seq.dynamic_decode(inference_decoder, maximum_iterations=self.max_decode_step) self.decoder_pred_decode = inference_decoder_output.predicted_ids self.decoder_pred_decode = tf.transpose(self.decoder_pred_decode,perm=[0, 2, 1]
请教一下，这里predicted_ids的维度应该是[batch_size, decoder_targets_length, beam_size]，为什么最后要做一次transpose呢？

[1033020837]

这个需要查阅Tensorflow相关的API，我是照着API写的