计算条件下的重复次数

Question

您好首先，我向您展示我的数据：

^{tb1}$

知道这些数据是可以重复的（这是我的情况所需要的，我想保留它们）

我想做的是一个专栏，告诉我每个人有多少个孩子（显然没有孩子的人会有na，但这没关系），所以它看起来像这样：

^{tb2}$

我怎样才能继续

警告！！！我没有在我的示例中提到这一点，但在我的数据中，这些行重复出现，我不必删除它们——想象一下Sara的另一行包含相同的信息

Answer 1

如果原始数据存储在数据帧（df）中，则可以使用：

df = df.replace('Na', pd.NA)
df['children'] = df['registered'].map(df.drop_duplicates(subset='registered')['Daughter of'].value_counts().to_dict())

Answer 2

您可以创建一个按每个“已注册”的“子项数”分组的辅助数据框，以便以后将其与原始数据框合并。其内容如下：

import pandas as pd

# Setting the data
all_rows = [["7D","Sara","8A"],
            ["8A","Rosa","Na"],
            ["4D","Jess","8A"],
            ["6B","Veronica","Na"],
            ["8L","Sophia","6B"],
            ["7N","iria","Na"],
            ["7D","Sara","8A"],
            ["8A","Rosa","Na"]] 

df = pd.DataFrame(all_rows, columns=["registered","name","daughter_of"])

# Df aux
df_grouped = df.drop_duplicates().groupby(["daughter_of"])["daughter_of"].count().reset_index(name="children")

# Renaming columns so the join is made correctly
df_grouped.columns = ["registered", "children"]

# Joining
df = pd.merge(df,df_grouped[df_grouped["registered"]!="Na"],on=["registered"],how='left')

这是我收到的输出

  registered      name daughter_of  children
0         7D      Sara          8A       NaN
1         8A      Rosa          Na       2.0
2         4D      Jess          8A       NaN
3         6B  Veronica          Na       1.0
4         8L    Sophia          6B       NaN
5         7N      iria          Na       NaN
6         7D      Sara          8A       NaN
7         8A      Rosa          Na       2.0

“已注册”字段只考虑一次行数