pandas 只替换正则表达式 [英] Pandas replace only working with regex

问题描述

我已经找到了解决方法，但是我仍然对正在发生的事情感到好奇:

I already found a workaround, but I'm still curious about what's going on:

当我尝试像这样在熊猫中进行替换时:

When I try to do replace in Pandas like so:

merged4[['column1', 'column2', 'column3']] = merged4[['column1', 'column2', 'column3']].replace(to_replace='.', value=',')

它不起作用.例如，我尝试了Inplace的所有不同变体.由于列为float64，因此我也做了astype(str).

It's not working. I tried all different variants with Inplace for example. I also did astype(str) since the columns were float64.

但是，当我这样做时:

merged4[['column1', 'column2', 'column3']] = merged4[['column1', 'column2', 'column3']].replace(to_replace='\.', value=',', regex=True)

它就像魅力一样运作.

It's working like charm.

有什么问题吗?

推荐答案

当您不使用regex=True时，该方法将查找replace_to值的完全匹配.当您使用regex=True时，它也会寻找子字符串.因此，当您使用该参数时，您的代码即可工作. 例子

When you don't use regex=True the method will look for the exact match of the replace_to value. When you use regex=True it will look for the sub strings too. So your code works when you use that parameter. Example

当您不使用regex=True参数时，replace方法将查找系列中replace_to值的精确匹配.

When you dont use the regex=True parameter, the replace method will look for the exact match of the replace_to value in the series.

df = pd.DataFrame()
df['k'] = ['YoYo.','mins.s.s','sos.s.','.','mama.mia']
df['k'].replace('.',',')

0       YoYo.
1    mins.s.s
2      sos.s.
3           ,
4    mama.mia

使用regex = True时，它甚至与系列的子字符串匹配并更改字符串

When you use regex = True it even matches the substrings of the series and changes the strings

df = pd.DataFrame()
df['k'] = ['YoYo.','mins.s.s','sos.s.','.','mama.mia']
df['k'].replace('\.',',',regex=True)

0       YoYo,
1    mins,s,s
2      sos,s,
3           ,
4    mama,mia

这篇关于 pandas 只替换正则表达式的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！