增加t5-copy模型,在t5-pegasus的基础上增加了pointer generator,用t5-pegasus的预训练任务继续训练 增加t5-copy-large模型,在t5-copy的基础上用公开的文本摘要数据集进行训练 增加examples,基于pytorch_lightning的多卡训练 数据集:AdvertiseGen modelbleurouge-1rouge-2rouge-2 ...
Construct a Pegasus tokenizer. Based on WordPiece.This tokenizer inherits from [`PreTrainedTokenizer`] which contains most of the main methods. Users should refer to this superclass for more information regarding those methods.Args: vocab_file (`str`): ...