我有这个数据框:
Dt1 Dt2
0 8/21/19 8/31/19
1 8/21/19 8/31/19
2 8/21/19 8/31/19
3 8/30/19 8/31/19
然后我写了这段代码:
for ind in df.index:
date_str1 = df['Dt1'][ind]
date_str2 = df['Dt2'][ind]
date_object1 = datetime.strptime(date_str1, " %m/%d/%y")
date_object2 = datetime.strptime(date_str2, " %m/%d/%y")
d = date_object2-date_object1
diff = d.days
print(diff)
我的结果是:
10
10
10
1
这是我对结果的期望,现在我要对数据帧中的每一行进行操作,我想创建一个新列(date_diff)并将这些结果添加到每一行,因此最后我有了像这样的东西:
Dt1 Dt2 Date_diff
0 8/21/19 8/31/19 10
1 8/21/19 8/31/19 10
2 8/21/19 8/31/19 10
3 8/30/19 8/31/19 1
答案 0 :(得分:5)
您可以简单地做到:
df['Dt1'] = pd.to_datetime(df['Dt1'])
df['Dt2'] = pd.to_datetime(df['Dt2'])
df['Date_diff'] = df['Dt2'].sub(df['Dt1']).dt.days
答案 1 :(得分:0)
要么创建一列(包含纯零,NaN或其他内容),然后将其写在循环的末尾,即df.iloc[ind,"Date_Diff"]=diff
或拥有列表Date_diff=[]
,然后附加在末尾{{ 1}}-最后将其写入您的df Date_diff.append(diff)