错误：无法解析位置 6116 处的字符串“*” - 将对象类型转换为 Int - Pandas
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/45177209/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Error: Unable to parse string "*" at position 6116 - Convert Object Type to Int - Pandas
提问by i.n.n.m
This question has been asked in many threads and has worked for others, but not for me. I am trying to convert objectdata type into intto perform a group by aggregation.
Following are what I tried and the errors I got so far, (I am using python 3)
According to this link, I tried these two:
这个问题已经在许多线程中被问到并且对其他人有用，但对我却没有。我正在尝试将object数据类型转换int为通过聚合执行组。以下是我尝试过的以及到目前为止遇到的错误（我正在使用 python 3）根据此链接，我尝试了以下两个：
df['my_var'] = df['my_var'].astype(str).astype(int)
df['my_var'] = df['my_var'].astype(int)
Same error for both:
两者都有相同的错误：
ValueError: invalid literal for int() with base 10: '*'
ValueError：int() 的无效文字，基数为 10：'*'
And then I tried,
然后我尝试了，
df['my_var'] = pd.to_numeric(df['my_var'])
I got an error:
我有一个错误：
ValueError: Unable to parse string "*" at position 6116
ValueError：无法解析位置 6116 处的字符串“*”
This is how dtypeslooks like,
长dtypes这样，
print (df.dtypes)
my_var object
dtype: object
I know some of the similar questions are down voted, however, I did not succeed using those answers. Is it a version error? I am finding it difficult to understand this error. Any help or suggestion would be appreciated.
我知道一些类似的问题被否决了，但是，我没有成功使用这些答案。是版本错误吗？我发现很难理解这个错误。任何帮助或建议将不胜感激。
采纳答案by i.n.n.m
After getting suggestions from #DYZ and #MaxU, it was an error due to the special character *in a row in in my DataFrame. (Error message was obvious)
从#DYZ 和#MaxU 获得建议后，由于*我的 DataFrame 中一行中的特殊字符，这是一个错误。（错误信息很明显）
As suggested, using,
按照建议，使用，
df[df['my_var']=='*']
and
和
df.loc[pd.to_numeric(df['my_var'], errors='coerce').isnull()]
I found where exactly the special character was. Then used regular expression method to strip off special characters using this thread.
我找到了特殊字符的确切位置。然后使用正则表达式方法使用此线程剥离特殊字符。
回答by A.Kot
I used 0 to replace anything that isn't a number but you can use any other value that makes sense to you e.g. -999999 (not a suggested practice obviously but just an example)
我用 0 来替换任何不是数字的东西，但你可以使用任何其他对你有意义的值，例如 -999999（显然不是建议的做法，只是一个例子）
pd.to_numeric(df.my_var, errors='coerce').fillna(0).astype(int)
