在 Panda 中使用 Add() 函数合并多个数据帧

Question

给定三个数据框，其中包含一些国家获得的金牌、银牌和铜牌数量，确定每个国家获得的奖牌总数。 注意：所有三个数据框的国家并不完全相同。另外，按照奖牌总数降序对最终数据框进行排序。

这是我下面的代码 - 但我没有得到想要的输出。有人可以提出问题吗？

import numpy as np 
import pandas as pd

# Defining the three dataframes indicating the gold, silver, and bronze medal counts
# of different countries
gold = pd.DataFrame({'Country': ['USA', 'France', 'Russia'],
                         'Medals': [15, 13, 9]}
                    )
silver = pd.DataFrame({'Country': ['USA', 'Germany', 'Russia'],
                        'Medals': [29, 20, 16]}
                    )
bronze = pd.DataFrame({'Country': ['France', 'USA', 'UK'],
                        'Medals': [40, 28, 27]}
                    )
#gold.set_index('Country',inplace = True)
#silver.set_index('Country',inplace = True)
#bronze.set_index('Country',inplace = True)

Total = gold.add(silver,fill_value = 0).add(bronze,fill_value = 0)
Total.sort_values('Medals',ascending = True)

Answer 1

你可以试试：

pd.concat([gold, silver, bronze]).groupby('Country').sum().\
          sort_values('Medals', ascending=False).reset_index()

如果您愿意，您可以将三个数据框合二为一。它按国家/地区分组，您将获得每个国家/地区的奖牌总和。最后我们按降序排序并重置索引。

输出：

    Country  Medals
0       USA      72
1    France      53
2        UK      27
3    Russia      25
4   Germany      20

Answer 2

你也可以这样做：

    gold.set_index('Country', inplace=True)
    silver.set_index('Country', inplace=True)
    bronze.set_index('Country', inplace=True)
    #print(gold)
    #print(silver)
    #print(bronze)

   Total= gold.add(silver,    fill_value=0).add(bronze,fill_value=0).sort_values('Medals', ascending=False)

输出：

             Medals
    Country        
    USA        72.0
    France     53.0
    UK         27.0
    Russia     25.0
    Germany    20.0

Answer 3

import numpy as np 
import pandas as pd

# Defining the three dataframes indicating the gold, silver, and bronze medal counts
# of different countries
gold = pd.DataFrame({'Country': ['USA', 'France', 'Russia'],
                         'Medals': [15, 13, 9]}
                    )
silver = pd.DataFrame({'Country': ['USA', 'Germany', 'Russia'],
                        'Medals': [29, 20, 16]}
                    )
bronze = pd.DataFrame({'Country': ['France', 'USA', 'UK'],
                        'Medals': [40, 28, 27]}
                    )

# Set the index of the dataframes to 'Country' so that you can get the countrywise
# medal count
gold.set_index('Country', inplace = True)
silver.set_index('Country', inplace = True) 
bronze.set_index('Country', inplace = True) 

# Add the three dataframes and set the fill_value argument to zero to avoid getting
# NaN values
total = gold.add(silver, fill_value = 0).add(bronze, fill_value = 0)

# Sort the resultant dataframe in a descending order
total = total.sort_values(by = 'Medals', ascending = False).astype("int64")

# Print the sorted dataframe
print(total)

int64 is used to convert the float value into integer and 64 indicates 64bit memory location

Answer 4

import numpy as np 
import pandas as pd

# Defining the three dataframes indicating the gold, silver, and bronze medal counts
# of different countries
gold = pd.DataFrame({'Country': ['USA', 'France', 'Russia'],
                         'Medals': [15, 13, 9]}
                    )
silver = pd.DataFrame({'Country': ['USA', 'Germany', 'Russia'],
                        'Medals': [29, 20, 16]}
                    )
bronze = pd.DataFrame({'Country': ['France', 'USA', 'UK'],
                        'Medals': [40, 28, 27]}
                    )
gold.set_index('Country' , inplace = True)
silver.set_index('Country' , inplace = True) 
bronze.set_index('Country' , inplace = True ) 

total = gold.add(silver , fill_value = 0).add(bronze , fill_value = 0)
total = total.sort_values(by = 'Medals', ascending = False)
total = total.astype(int)
print(total)