Pandas 基于日期列筛选数据框
我有一个熊猫数据框,如下所示Pandas 基于日期列筛选数据框,pandas,datetime,pandas-groupby,Pandas,Datetime,Pandas Groupby,我有一个熊猫数据框,如下所示 Unit_ID Added_Date Status 105 2019-10-02 07:18:18 Rented 106 2020-15-01 07:18:17 Rented 105 2019-10-02 07:18:19 Rented 108 2020-15-01 07:18:18
Unit_ID Added_Date Status
105 2019-10-02 07:18:18 Rented
106 2020-15-01 07:18:17 Rented
105 2019-10-02 07:18:19 Rented
108 2020-15-01 07:18:18 Vacant
根据以上内容,我想了解在过去10天内根据日期列添加的单位ID
预期产出:
Unit_ID Added_Date Status
106 2020-15-01 07:18:17 Rented
108 2020-15-01 07:18:18 Vacant
以下是一种方法:
today = pd.to_datetime('today')
n = 10 # last n days
filter_criteria = df['Added_Date'].sub(today).abs().apply(lambda x: x.days <= n)
df.loc[filter_criteria]
Unit_ID Added_Date Status
106 106 2020-01-15 07:18:17 Rented
108 108 2020-01-15 07:18:18 Vacant
today=pd.to\u datetime('today'))
n=10#最后n天
filter_criteria=df['Added_Date'].sub(today).abs().apply(lambda x:x.days您还可以使用.dt.days
访问器并与10:
这里有另一种使用pd.DateOffset
from datetime import datetime
df.loc[df['Added_Date'] >= (datetime.today() - pd.DateOffset(days=10))]
Unit_ID Added_Date Status
1 106 2020-01-15 07:18:17 Rented
3 108 2020-01-15 07:18:18 Vacant
from datetime import datetime
df.loc[df['Added_Date'] >= (datetime.today() - pd.DateOffset(days=10))]
Unit_ID Added_Date Status
1 106 2020-01-15 07:18:17 Rented
3 108 2020-01-15 07:18:18 Vacant