Python 在列中附加值为_counts（）的列

我有一个名为output的数据帧，如下所示：

   created_at
0  1/8/2017 0:00
1  1/8/2017 0:00
2  1/8/2017 0:00
3  1/8/2017 0:00
4  1/8/2017 0:00
5  1/8/2017 1:00
6  1/8/2017 2:00
7  1/8/2017 3:00

1/8/2017 0:00    5
1/8/2017 1:00    1
1/8/2017 3:00    1
1/8/2017 2:00    1

   created_at        count
0  1/8/2017 0:00     NaN
5  1/8/2017 1:00     NaN
6  1/8/2017 2:00     NaN
7  1/8/2017 3:00     NaN

import pandas as pd

df1 = pd.read_csv(path1)
df2 = pd.read_csv(path2)
output = pd.merge(df1, df2, how="inner", on="created_at")
df3 = output.created_at.value_counts()

output = output.drop_duplicates()
output = output.sort_values(by=['created_at'])
output['count'] = df3


print(output,'\n\n')

我想计算特定时间在名为df3的数据帧中出现的次数。结果如下：

   created_at
0  1/8/2017 0:00
1  1/8/2017 0:00
2  1/8/2017 0:00
3  1/8/2017 0:00
4  1/8/2017 0:00
5  1/8/2017 1:00
6  1/8/2017 2:00
7  1/8/2017 3:00

1/8/2017 0:00    5
1/8/2017 1:00    1
1/8/2017 3:00    1
1/8/2017 2:00    1

   created_at        count
0  1/8/2017 0:00     NaN
5  1/8/2017 1:00     NaN
6  1/8/2017 2:00     NaN
7  1/8/2017 3:00     NaN

import pandas as pd

df1 = pd.read_csv(path1)
df2 = pd.read_csv(path2)
output = pd.merge(df1, df2, how="inner", on="created_at")
df3 = output.created_at.value_counts()

output = output.drop_duplicates()
output = output.sort_values(by=['created_at'])
output['count'] = df3


print(output,'\n\n')

我想在df3中添加两个标题，分别为created_at和count

我首先做的是从输出数据帧中删除重复项，并对值进行排序，得到如下结果：

   created_at
0  1/8/2017 0:00
5  1/8/2017 1:00
6  1/8/2017 2:00
7  1/8/2017 3:00

现在，我在输出数据帧中添加了列count，但得到的结果如下：

   created_at
0  1/8/2017 0:00
1  1/8/2017 0:00
2  1/8/2017 0:00
3  1/8/2017 0:00
4  1/8/2017 0:00
5  1/8/2017 1:00
6  1/8/2017 2:00
7  1/8/2017 3:00

1/8/2017 0:00    5
1/8/2017 1:00    1
1/8/2017 3:00    1
1/8/2017 2:00    1

   created_at        count
0  1/8/2017 0:00     NaN
5  1/8/2017 1:00     NaN
6  1/8/2017 2:00     NaN
7  1/8/2017 3:00     NaN

import pandas as pd

df1 = pd.read_csv(path1)
df2 = pd.read_csv(path2)
output = pd.merge(df1, df2, how="inner", on="created_at")
df3 = output.created_at.value_counts()

output = output.drop_duplicates()
output = output.sort_values(by=['created_at'])
output['count'] = df3


print(output,'\n\n')

我想要实现的是一个名为result的数据帧，它应该如下所示：

   created_at        count
0  1/8/2017 0:00     5
5  1/8/2017 1:00     1
6  1/8/2017 2:00     1
7  1/8/2017 3:00     1

我该怎么做？我的代码如下：

   created_at
0  1/8/2017 0:00
1  1/8/2017 0:00
2  1/8/2017 0:00
3  1/8/2017 0:00
4  1/8/2017 0:00
5  1/8/2017 1:00
6  1/8/2017 2:00
7  1/8/2017 3:00

1/8/2017 0:00    5
1/8/2017 1:00    1
1/8/2017 3:00    1
1/8/2017 2:00    1

   created_at        count
0  1/8/2017 0:00     NaN
5  1/8/2017 1:00     NaN
6  1/8/2017 2:00     NaN
7  1/8/2017 3:00     NaN

import pandas as pd

df1 = pd.read_csv(path1)
df2 = pd.read_csv(path2)
output = pd.merge(df1, df2, how="inner", on="created_at")
df3 = output.created_at.value_counts()

output = output.drop_duplicates()
output = output.sort_values(by=['created_at'])
output['count'] = df3


print(output,'\n\n')

任何和所有的帮助都将不胜感激

---

谢谢

在调用

值\u计数

后，使用

重置\u索引

重命名\u轴

df.created_at.value_counts().rename_axis('created_at').reset_index(name='count')

      created_at  count
0  1/8/2017 0:00      5
1  1/8/2017 2:00      1
2  1/8/2017 1:00      1
3  1/8/2017 3:00      1

---

或者，使用

groupby

+

agg

：

df.groupby('created_at').created_at.agg([('count', 'count')]).reset_index()

      created_at  count
0  1/8/2017 0:00      5
1  1/8/2017 1:00      1
2  1/8/2017 2:00      1
3  1/8/2017 3:00      1

---

结果df不是我所拥有的，而是我所期望的输出