pivot_table dropna 参数未按预期工作

Question

如果 True 和 False 产生相同的结果，那么 dropna 参数有什么用？

import pandas as pd
import seaborn as sns

# Load the dataset
df = sns.load_dataset("penguins")

df=df.melt( id_vars=['species','island','sex'],
         value_vars=['bill_length_mm','bill_depth_mm','flipper_length_mm','body_mass_g'], 
         var_name='metric', 
         value_name='value'
       )

k1 = df.pivot_table(index='island', columns='metric', values='value',
                                   aggfunc=['mean', 'count','size'], dropna=False,observed=False).reset_index()
k2 = df.pivot_table(index='island', columns='metric', values='value',
                                   aggfunc=['mean', 'count','size'], dropna=True,observed=False).reset_index()

print("Pivot table with dropna=False:\n")
print(k1)
print("\n\nPivot table with dropna=True:\n")
print(k2)

输出：

Pivot table with dropna=False:

           island          mean                                                \
metric            bill_depth_mm bill_length_mm  body_mass_g flipper_length_mm   
0          Biscoe     15.874850      45.257485  4716.017964        209.706587   
1           Dream     18.344355      44.167742  3712.903226        193.072581   
2       Torgersen     18.429412      38.950980  3706.372549        191.196078   

               count                                               \
metric bill_depth_mm bill_length_mm body_mass_g flipper_length_mm   
0                167            167         167               167   
1                124            124         124               124   
2                 51             51          51                51   

                size                                               
metric bill_depth_mm bill_length_mm body_mass_g flipper_length_mm  
0                168            168         168               168  
1                124            124         124               124  
2                 52             52          52                52  


Pivot table with dropna=True:

           island          mean                                                \
metric            bill_depth_mm bill_length_mm  body_mass_g flipper_length_mm   
0          Biscoe     15.874850      45.257485  4716.017964        209.706587   
1           Dream     18.344355      44.167742  3712.903226        193.072581   
2       Torgersen     18.429412      38.950980  3706.372549        191.196078   

               count                                               \
metric bill_depth_mm bill_length_mm body_mass_g flipper_length_mm   
0                167            167         167               167   
1                124            124         124               124   
2                 51             51          51                51   

                size                                               
metric bill_depth_mm bill_length_mm body_mass_g flipper_length_mm  
0                168            168         168               168  
1                124            124         124               124  
2                 52             52          52                52  

Answer 1

import pandas as pd
import seaborn as sns
import numpy as np

# Load the dataset
df = sns.load_dataset("penguins")



# Reshape the dataframe
df = df.melt(id_vars=['species', 'island', 'sex'],
             value_vars=['bill_length_mm', 'bill_depth_mm', 'flipper_length_mm', 'body_mass_g'], 
             var_name='metric', 
             value_name='value')

df=df.dropna(subset='value')

# Introduce some missing values
df.loc[0:3, 'island'] = np.nan
df.loc[2:4, 'metric'] = np.nan

# Create pivot tables with dropna=False and dropna=True
pivot_with_margins = df.pivot_table(index='island', columns='metric', values='value',
                                    aggfunc=['size','count'], dropna=False)

pivot_without_margins = df.pivot_table(index='island', columns='metric', values='value',
                                    aggfunc=['size','count'], dropna=True)

print("Pivot table with margins (dropna=False):")
print(pivot_with_margins)

print("\n\nPivot table with margins (dropna=True):")
print(pivot_without_margins)


Pivot table with margins (dropna=False):
                   size                                                    \
metric    bill_depth_mm bill_length_mm body_mass_g flipper_length_mm  NaN   
island                                                                      
Biscoe            167.0          167.0       167.0             167.0  NaN   
Dream             124.0          124.0       124.0             124.0  NaN   
Torgersen          51.0           47.0        51.0              51.0  1.0   
NaN                 NaN            2.0         NaN               NaN  1.0   

                  count                                                    
metric    bill_depth_mm bill_length_mm body_mass_g flipper_length_mm  NaN  
island                                                                     
Biscoe            167.0          167.0       167.0             167.0  NaN  
Dream             124.0          124.0       124.0             124.0  NaN  
Torgersen          51.0           47.0        51.0              51.0  1.0  
NaN                 NaN            2.0         NaN               NaN  1.0  


Pivot table with margins (dropna=True):
                   size                                               \
metric    bill_depth_mm bill_length_mm body_mass_g flipper_length_mm   
island                                                                 
Biscoe              167            167         167               167   
Dream               124            124         124               124   
Torgersen            51             47          51                51   

                  count                                               
metric    bill_depth_mm bill_length_mm body_mass_g flipper_length_mm  
island                                                                
Biscoe              167            167         167               167  
Dream               124            124         124               124  
Torgersen            51             47          51                51