日常报错问题

1. could not determine the shape of object type ‘Series’

解：X_test = torch.Tensor(X_test.to_numpy())

02.TypeError: ‘numpy.float64’ object cannot be interpreted as an integer

解：int(报错的变量)

03. Expected tensor for argument #1 ‘indices’ to have scalar type Long；but got torch.IntTensor instead

解：出错的变量.long() # 将报错位置处的变量进行type转换

04. doc2bow expects an array of unicode tokens on input, not a single string

解：是在使用LDA模型时出现的错误，dictionary = corpora.Dictionary(words_list)现有代码的情况下将 wordlist改为[wordlist]

05. 安装docx模块出现Import Error: No module named 'exceptions’的解决方案

解：安装 pip install python-docx

06. ‘gbk‘ codec can‘t decode byte 0x80 in position 309: illegal multibyte sequence

解：添加encoding for line in open("xxx.txt","r",encoding='UTF-8'):

07.np.nan is an invalid document, expected byte or unicode string.

解：输入数据有误。详见：https://blog.csdn.net/qq_35182128/article/details/118578199