跳转至

Python | Pandas pants.rtruediv()

原文:https://www.geeksforgeeks.org/python-pandas-panel-rtruediv/

在熊猫中,面板是一个非常重要的三维数据容器。三个轴的名称旨在为描述涉及面板数据的操作,特别是面板数据的计量经济学分析提供一些语义含义。

熊猫**Panel.rtruediv()**功能用于获取系列和数据框/面板的浮动划分。

语法: Panel.rtruediv(其他,轴=0)

参数: 其他:数据框或面板 轴:轴广播结束

返回:面板

代码#1:

# importing pandas module 
import pandas as pd 
import numpy as np 

df1 = pd.DataFrame({'a': ['Geeks', 'For', 'geeks', 'real'], 
                    'b': [111, 123, 425, 1333]}) 

df2 = pd.DataFrame({'a': ['I', 'am', 'dataframe', 'two'], 
                    'b': [100, 100, 100, 100]}) 

data = {'item1':df1, 'item2':df2}

# creating Panel 
panel = pd.Panel.from_dict(data, orient ='minor') 
print("panel['b'] is - \n\n", panel['b']) 

print("\nFloating Dividing panel['b'] with df2['b'] using rtruediv() method - \n") 
print("\n", panel['b'].rtruediv(df2['b'], axis = 0)) 

Output:

panel['b'] is - 

    item1  item2
0    111    100
1    123    100
2    425    100
3   1333    100

Floating Dividing panel['b'] with df2['b'] using rtruediv() method - 

       item1  item2
0  0.900901      1
1  0.813008      1
2  0.235294      1
3  0.075019      1

代码#2:

# importing pandas module 
import pandas as pd 
import numpy as np 

df1 = pd.DataFrame({'a': ['Geeks', 'For', 'geeks', 'for', 'real'], 
                    'b': [11, 1.025, 333, 114.48, 1333]}) 

data = {'item1':df1, 'item2':df1} 

# creating Panel 
panel = pd.Panel.from_dict(data, orient ='minor') 
print("panel['b'] is - \n\n", panel['b'], '\n') 

# Create a 5 * 5 dataframe 
df2 = pd.DataFrame(np.random.rand(5, 2), columns =['item1', 'item2']) 
print("Newly create dataframe with random values is - \n\n", df2)

print("\nFloating Dividing panel['b'] with df2 using rtruediv() method - \n") 
print(panel['b'].rtruediv(df2, axis = 0)) 

Output:

panel['b'] is - 

       item1     item2
0    11.000    11.000
1     1.025     1.025
2   333.000   333.000
3   114.480   114.480
4  1333.000  1333.000 

Newly create dataframe with random values is - 

       item1     item2
0  0.809990  0.358019
1  0.765535  0.101660
2  0.788052  0.742361
3  0.964031  0.618767
4  0.862207  0.131611

Floating Dividing panel['b'] with df2 using rtruediv() method - 

      item1     item2
0  0.073635  0.032547
1  0.746864  0.099180
2  0.002367  0.002229
3  0.008421  0.005405
4  0.000647  0.000099

代码#3:

# importing pandas module 
import pandas as pd 
import numpy as np 

df1 = pd.DataFrame({'a': ['Geeks', 'For', 'geeks', 'for', 'real'], 
                    'b': [11, 1.025, 333, 114.48, 1333]}) 

df2 = pd.DataFrame({'a': ['I', 'am', 'DataFrame', 'number', 'two'], 
                    'b': [10, 10, 10, 110, 110]})                     

data = {'item1':df1, 'item2':df2} 

# creating Panel 
panel = pd.Panel.from_dict(data, orient ='minor') 

print("panel['b'] is - \n\n", panel['b'], '\n') 

print("\nFloating Dividing panel['b']['item1'] with df2['b'] or panel['b']['item2'] using rtruediv() method - \n") 
print("\n", panel['b']['item1'].rtruediv(df2['b'], axis = 0)) 

Output:

panel['b'] is - 

       item1  item2
0    11.000     10
1     1.025     10
2   333.000     10
3   114.480    110
4  1333.000    110 

Floating Dividing panel['b']['item1'] with df2['b'] or panel['b']['item2'] using rtruediv() method - 

 0    0.909091
1    9.756098
2    0.030030
3    0.960867
4    0.082521
dtype: float64



回到顶部