明明已经对超过512长度的序列做了处理，但这个怎么都处理不了，经过检查发现在对子句之间添加[SEP]标记时，错误的将[SEP]写成了[sep]，致使电脑在识别的时候无法将[sep]识别出来，而是识别成了[ , s , e , p , ] 这几个分开的，所以长度比正常的长而且代码也无法精简到512（虽然已经精简了但是还是缩不到512，这个后期有待进一步研究，目前原因仍不清楚）
免责声明！
本站转载的文章为个人学习借鉴使用，本站对版权不负任何法律责任。如果侵犯了您的隐私权益，请联系本站邮箱yoyou2525@163.com删除。
猜您在找
RuntimeError: The size of tensor a (30) must match the size of tensor b (54) at non-singleton dimension 0
RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 1 #805
the-trustanchors-parameter-must-be-non-empty
has invalid type , must be a string or Tensor
SELECT DISTINCT ON expressions must match initial ORDER BY expressions
InvalidArgumentError: You must feed a value for placeholder tensor 'Placeholder_1' with dtype float and shape [?,10]
RuntimeError:stack expects each tensor to be equal size,but got [3,224,244] at entry 0 and [1,224,244] at entry 7
关于EF分页查询报错（Count must have a non-negative value.）的解决方案
C++ Error : initial value of reference to non-const must be an lvalue
nginx报警：nginx: [warn] could not build optimal server_names_hash, you should increase either server_names_hash_max_size: 512 or server_names_hash_bucket_size: 64; ignoring server_names_hash_bucket_size