Python对值进行排序并根据唯一键对它们进行分组

Question

我有一个元组元素列表，如下所示。我想将元素分组为多维行和列。例如：

说列表是“列表”：

[("Adam", "DeltaAir"),
("Bianca", "AlaskanAir"),
("Romeo", "DeltaAir"),
("Danaerys", "DragonAir"),
("Jon", "DragonAir"),
("Walter", "AlaskanAir")]

我想将此列表打印为：

------------------------------------------
Name  | AlaskanAir | DeltaAir | DragonAir
------------------------------------------
Adam                    *
Bianca      *
Romeo                   *
Danaerys                            *
Jon                                 *
Walter      *
------------------------------------------

我开始的目的是首先找到我想要作为行标题的所有唯一元素。

    row=[]
    for i in list:
        row.append(i[1])
    row = list(set(row))

然后我会遍历“行”中的元素，然后建立表格。我怎样才能轻松构建它？ 谢谢！

Answer 1

我们可以使用pandas：

import pandas as pd

df = pd.DataFrame([("Adam", "DeltaAir"),
("Bianca", "AlaskanAir"),
("Romeo", "DeltaAir"),
("Danaerys", "DragonAir"),
("Jon", "DragonAir"),
("Walter", "AlaskanAir")], columns=['name', 'value'])

result = pd.get_dummies(df, columns=['value']).rename(columns={f'value_{col}': col for col in df['value'].unique()}).replace({col: {0: '', 1: '*'} for col in df['value'].unique()})

print(result)

输出：

       name AlaskanAir DeltaAir DragonAir
0      Adam                   *          
1    Bianca          *                   
2     Romeo                   *          
3  Danaerys                             *
4       Jon                             *
5    Walter          *                   

这会将对应于每个个体的值转换为相关列中的 1 或 0。然后，我们只需将 1 替换为 * 并将 0 替换为空字符串。

注意pandas对于逻辑不是必须的，可以很简单，但是方便表格的对齐。

Answer 2

lst = [("Adam", "DeltaAir"),
("Bianca", "AlaskanAir"),
("Romeo", "DeltaAir"),
("Danaerys", "DragonAir"),
("Jon", "DragonAir"),
("Walter", "AlaskanAir")]

#Create pandas DataFrame with the names from the list
df = pd.DataFrame([elem[0] for elem in lst], columns=["Name"])
#Iterate over a set (unique values) of character properties (DeltaAir, AlaskanAir, DragonAir) 
for elem in set([elem[1] for elem in lst]):
    #make a list containing spaces or *'s for every character in the list
    #depending on the property we are just iterating over and add that list 
    # as a column to the DataFrame
    df[elem] = ["*" if item[1] == elem else " " for item in lst] 

编辑您的评论：

您可以使用 groupby 和聚合按名称组合值（如果这不是您的意思，请澄清）。

df.reindex(sorted(df.columns))
df2 = pd.DataFrame(sorted(list(df["Name"].unique())), columns = ["Name"])
for elem in set([elem[1] for elem in lst]):   
    df2[elem] = list(df.groupby(['Name'])[elem].agg(lambda x: "*" if "*" in x.values else " "))

添加更多信息

谢谢弗洛里安。我的意思是如果有像下面这样的重复名称，它应该适当地填充相应的航空公司行。例如：亚当和罗密欧出现两次就像，而不是同名有两个单独的行。

[("Adam", "DeltaAir"),
("Bianca", "AlaskanAir"),
("Romeo", "DeltaAir"),
("Danaerys", "DragonAir"),
("Jon", "DragonAir"),
("Walter", "AlaskanAir"),
("Adam", "AlaskanAir"),
("Romeo", "DragonAir")]

------------------------------------------
Name  | AlaskanAir | DeltaAir | DragonAir
------------------------------------------
Adam        *           *
Bianca      *
Romeo                   *           *
Danaerys                            *
Jon                                 *
Walter      *
------------------------------------------