search

shiyybua / NER

基于tensorflow深度学习的中文的命名实体识别
1.04k stars 400 forks source link

ValueError: could not broadcast input array from shape (299) into shape (300) #28

Open xiaoheiiii opened 6 years ago

xiaoheiiii commented 6 years ago

请问在train时一直报这个错误 embeddings[index] = coefs # 将词和对应的向量存到字典里 ValueError: could not broadcast input array from shape (299) into shape (300) 这要怎么解决。

maggiegyh commented 6 years ago

I have the same problem, at begging the error is about 'gbk' ,then I change 'utf-8', now the error like the description above

zhangmingnine commented 6 years ago

取出词向量的时候分割出了问题吧

szq1292718561 commented 6 years ago

是哪里出问题了吗，我也是这种情况

zcoo commented 5 years ago

同问

RoacherM commented 5 years ago

GLOVE has the same problem,it seems that there is something wrong in some lines of wiki.zh.vec . https://github.com/zhiguowang/BiMPM/issues/23

huanghonggit commented 5 years ago

I also encountered this problem, how can I solve it in the end?

LiuShaohan commented 5 years ago

这个问题，到底怎么解决的？有大佬回复下吗?

urextra commented 3 years ago

各位大佬，解决了吗