Python|熊猫 Panel.rfloordiv()
在 Pandas 中,Panel 是一个非常重要的 3D 数据容器。 3 个轴的名称旨在为描述涉及面板数据的操作提供一些语义含义,特别是面板数据的计量经济学分析。
在 Pandas Panel.rfloordiv()
函数用于获取系列和数据框/面板的整数除法。
Syntax: Panel.rfloordiv(other, axis=0)
Parameters:
other : DataFrame or Panel
axis : Axis to broadcast over
Returns: Panel
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': [5, 5, 2, 10]})
data = {'item1':df1, 'item2':df2}
# creating Panel
panel = pd.Panel.from_dict(data, orient ='minor')
print("panel['b'] is - \n\n", panel['b'])
print("\nInteger Dividing panel['b'] with df2['b'] using rfloordiv() method - \n")
print("\n", panel['b'].rfloordiv(df2['b'], axis = 0))
输出:
panel['b'] is -
item1 item2
0 111 5
1 123 5
2 425 2
3 1333 10
Integer Dividing panel['b'] with df2['b'] using rfloordiv() method -
item1 item2
0 0 1
1 0 1
2 0 1
3 0 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]})
df2 = pd.DataFrame({'a': ['I', 'am', 'DataFrame', 'number', 'two'],
'b': [3, 3, 3, 13, 27]})
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("\nInteger Dividing panel['b']['item1'] with df2['b'] or panel['b']['item2'] using rfloordiv() method - \n")
print("\n", panel['b']['item1'].rfloordiv(df2['b'], axis = 0))
输出:
panel['b'] is -
item1 item2
0 11.000 3
1 1.025 3
2 333.000 3
3 114.480 13
4 1333.000 27
Integer Dividing panel['b']['item1'] with df2['b'] or panel['b']['item2'] using rfloordiv() method -
0 0
1 2
2 0
3 0
4 0
dtype: float64