如果字符串存在于列表中，则替换整个数据框中的字符串答案

【问题标题】：Replace strings in entire dataframe if they are present in a list如果字符串存在于列表中，则替换整个数据框中的字符串
【发布时间】：2017-09-09 16:29:23
【问题描述】：

感谢您花时间访问我的帖子。我有以下数据框：

df1
        col1                                    col2
    1   virginia is cold, canada is cold too    virginia is cold, canada is cold too
    2   florida, virginia, washington are good  florida, virginia, washington are good
    3   georgia, alabama, virginia are hot      virginia is cold, canada is cold too
    4   virginia, ohio, new castle are great    hawaii, nebreska is wonderful
    5   hawaii, nebreska is wonderful           virginia, ohio, new castle are great

另外，我有一个包含字符串的列表：

lst = ['virginia', 'hot', 'too']

如果它与列表中的一个字符串匹配，我想用“xxxxxx”替换整个数据帧中的字符串。例如，替换后我的数据框将如下所示：

 df1
            col1                                    col2
        1   xxxxxx is cold, canada is cold xxxxxx   xxxxxx is cold, canada is cold xxxxxx
        2   florida, xxxxxx, washington are good    florida, xxxxxx, washington are good
        3   georgia, alabama, xxxxxx are xxxxxx     xxxxxx is cold, canada is cold xxxxxx
        4   xxxxxx, ohio, new castle are great      hawaii, nebreska is wonderful
        5   hawaii, nebreska is wonderful           xxxxxx, ohio, new castle are great

到目前为止，我已经尝试过，但它不起作用：

df1 = df1.replace(lst, "xxxxxx")

【问题讨论】：

标签： python list pandas dataframe replace

【解决方案1】：

尝试遍历列表lst，如下所示：

import pandas as pd

...
lst = ['virginia', 'hot', 'too']
for s in lst:
    df1.replace(s, 'xxxxx', inplace=True)

print( df1)

【讨论】：

【解决方案2】：

df1.replace(lst, 'x' * 5, regex=True)

                                  col1                                 col2
1  xxxxx is cold, canada is cold xxxxx  xxxxx is cold, canada is cold xxxxx
2  florida, xxxxx, washington are good  florida, xxxxx, washington are good
3    georgia, alabama, xxxxx are xxxxx  xxxxx is cold, canada is cold xxxxx
4    xxxxx, ohio, new castle are great        hawaii, nebreska is wonderful
5        hawaii, nebreska is wonderful    xxxxx, ohio, new castle are great

【讨论】：

【解决方案3】：

您可以从单词列表中创建字典并使用 regex：

lst = ['virginia', 'hot', 'too']
df1.replace({w: "xxxxxx" for w in lst}, regex=True)

【讨论】：