c2165224d198450a3b4329ae099a772aa65d51c5,fairseq/models/levenshtein_transformer.py,LevenshteinTransformerModel,forward_decoder,#LevenshteinTransformerModel#Any#Any#Any#Any#,318

Before Change

        if max_ratio is None:
            max_lens = output_tokens.new(output_tokens.size(0)).fill_(255)
        else:
            max_lens = (
                (~encoder_out["encoder_padding_mask"]).sum(1) * max_ratio
            ).clamp(min=10)

        // delete words
        // do not delete tokens if it is <s> </s>
        can_del_word = output_tokens.ne(self.pad).sum(1) > 2

After Change

            max_lens = output_tokens.new().fill_(255)
        else:
            if encoder_out["encoder_padding_mask"] is None:
                max_src_len = encoder_out["encoder_out"].size(1)
                src_lens = encoder_out["encoder_out"].new(bsz).fill_(max_src_len)
            else:
                src_lens = (~encoder_out["encoder_padding_mask"]).sum(1)
            max_lens = (src_lens * max_ratio).clamp(min=10).long()

        // delete words
        // do not delete tokens if it is <s> </s>
        can_del_word = output_tokens.ne(self.pad).sum(1) > 2

Instances