Python 如何获取基于日期时间的值计数

我编写了以下代码，创建了两个数据帧

nq

和

cmnt

nq

包含

UserId

和获得徽章的相应时间

日期

cmnt

包含

OwnerUserId

和用户发表评论的时间

CreationDate

我想统计每个用户在获得徽章1周之前和之后的所有日子里的评论数量，这样我就可以从中创建一个时间序列线图

下面的代码执行相同的操作，但对一部分数据产生错误，而对另一部分数据工作正常。请为我提供执行此任务的替代方法

nq

 UserId |   date 
     1      2009-10-17 17:38:32.590
     2      2009-10-19 00:37:23.067
     3      2009-10-20 08:37:14.143
     4      2009-10-21 18:07:51.247
     5      2009-10-22 21:25:24.483

cmnt

OwnerUserId | CreationDate
1             2009-10-16 17:38:32.590
1             2009-10-18 17:38:32.590
2             2009-10-18 00:37:23.067
2             2009-10-17 00:37:23.067
2             2009-10-20 00:37:23.067
3             2009-10-19 08:37:14.143
4             2009-10-20 18:07:51.247
5             2009-10-21 21:25:24.483

UserId     |   date                 |-7|-6|-5|-4|-3|-2|-1|0 |1 |2 |3 |4 |5 |6 |7
     1      2009-10-17 17:38:32.590 |0 |0 |0 |0 |0 |0 |1 |0 |1 |0 |0 |0 |0 |0 |0  
     2      2009-10-19 00:37:23.067 |0 |0 |0 |0 |0 |1 |1 |0 |1 |0 |0 |0 |0 |0 |0    
     3      2009-10-20 08:37:14.143 |0 |0 |0 |0 |0 |0 |1 |0 |0 |0 |0 |0 |0 |0 |0 
     4      2009-10-21 18:07:51.247 |0 |0 |0 |0 |0 |0 |1 |0 |0 |0 |0 |0 |0 |0 |0 
     5      2009-10-22 21:25:24.483 |0 |0 |0 |0 |0 |0 |1 |0 |0 |0 |0 |0 |0 |0 |0 

代码

t = pd.merge(nq, cmnt, left_on="UserId", right_on = "OwnerUserId")
t["days_diff"] = (t["CreationDate"] - t["date"]).dt.days
t["count"] = t.groupby(["UserId", "days_diff"]).OwnerUserId.transform("count")

all_days = pd.DataFrame(itertools.product(t.UserId.unique(), range(-7, 8)), )
all_days.columns = ["UserId", "day"]

t = pd.merge(t, all_days, left_on=["UserId", "days_diff"], right_on=["UserId", "day"], how = "right")
t = pd.pivot_table(t, index="UserId", columns="day", values="count", dropna=False)

res = pd.merge(nq, t, left_on="UserId", right_index=True)

print(res)

预期产出

OwnerUserId | CreationDate
1             2009-10-16 17:38:32.590
1             2009-10-18 17:38:32.590
2             2009-10-18 00:37:23.067
2             2009-10-17 00:37:23.067
2             2009-10-20 00:37:23.067
3             2009-10-19 08:37:14.143
4             2009-10-20 18:07:51.247
5             2009-10-21 21:25:24.483

UserId     |   date                 |-7|-6|-5|-4|-3|-2|-1|0 |1 |2 |3 |4 |5 |6 |7
     1      2009-10-17 17:38:32.590 |0 |0 |0 |0 |0 |0 |1 |0 |1 |0 |0 |0 |0 |0 |0  
     2      2009-10-19 00:37:23.067 |0 |0 |0 |0 |0 |1 |1 |0 |1 |0 |0 |0 |0 |0 |0    
     3      2009-10-20 08:37:14.143 |0 |0 |0 |0 |0 |0 |1 |0 |0 |0 |0 |0 |0 |0 |0 
     4      2009-10-21 18:07:51.247 |0 |0 |0 |0 |0 |0 |1 |0 |0 |0 |0 |0 |0 |0 |0 
     5      2009-10-22 21:25:24.483 |0 |0 |0 |0 |0 |0 |1 |0 |0 |0 |0 |0 |0 |0 |0 

此处，

-1

列表示在获得徽章前一天发表的评论，

1

列表示在获得徽章后一天发表的评论，依此类推

错误

ValueError: Length mismatch: Expected axis has 0 elements, new values have 2 elements

注意

---

错误是由这行代码引起的：

all_days.columns=[“UserId”，“day”]

这是否回答了您的问题？