new_tags in _score_sentence

[zhujiangang]

Can you explain a little bit more about the following code? Why need to do this?

        new_tags = autograd.Variable(torch.LongTensor(batch_size, seq_len))
        if self.gpu:
            new_tags = new_tags.cuda()
        for idx in range(seq_len):
            if idx == 0:
                ## start -> first score
                new_tags[:,0] =  (tag_size - 2)*tag_size + tags[:,0]

            else:
                new_tags[:,idx] =  tags[:,idx-1]*tag_size + tags[:,idx]