我的数据框有值：
data_df
0 student
1 sample text
2 student
3 no students
4 sample texting
5 random sample
我使用正则表达式提取带有单词'student'的行，我的结果如下：
regexdf
0 student
2 student
我的目标是在主数据框中创建一个具有 0 和 1 值的新列。即第 0 行应为 1，第 5 行应为零。（因为“regexdf”在第 0 行和第 2 行中有“学生”）如何匹配两者中的索引并创建列？
【问题讨论】：
标签： python regex pandas dataframe
使用正则表达式：
data_df = data_df.assign(regexdf = data_df[1].str.extract(r'(student)\b', expand=False))
data_df['student'] = data_df['regexdf'].notnull().mul(1)
print(data_df)
输出：
1 regexdf student
0 student student 1
1 sample text NaN 0
2 student student 1
3 no students NaN 0
4 sample texting NaN 0
5 random sample NaN 0
df_out = data_df.join(regexdf, rsuffix='regex')
df_out['pattern'] = df_out['1regex'].notnull().mul(1)
df_out['Count_Pattern'] = df_out['pattern'].cumsum()
print(df_out)
输出：
1 1regex pattern Count_Pattern
0 student student 1 1
1 sample text NaN 0 1
2 student student 1 2
3 no students NaN 0 2
4 sample texting NaN 0 2
5 random sample NaN 0 2
【讨论】：
def regex(series, regex): series = series.str.extract(regex) series1 = series.dropna() return (series1) regexdf = regex(comp_df,(r'.*(?<!no )(student ?in|student ?out).*(hallway|library)))
data_df.join(regexdf, how='outer') 或 data_df.merge(regexdf, how='outer', left_index=True, right_index=True)
你也可以这样做
df['bool'] = df[1].eq('student').astype(int)
或
df['bool'] = df[1].str.match(r'(student)\b').astype(int)
1 bool
0 student 1
1 sample text 0
2 student 1
3 no students 0
4 sample texting 0
5 random sample 0
如果你想要一个新的数据框，那么
ndf = df[df[1].eq('student')].copy()
【讨论】：