| pd.MultiIndex |
| 构造器 |
MI |
levels |
每个级别不重复的标签 |
| labels |
每个级别的整数指定每个位置 |
| *sortorder=None |
|
| *names=None |
每个级别的name |
| copy=False |
|
| verify_integrity=True |
检查levels和labels |
| .from_arrays() |
MI |
arrays ** |
list_like: [[一级索引],[二级索引]], 每一级长度都必须跟索引长度相同 |
| .from_product() |
MI |
iterable ** |
[无重复值的迭代器], 排列组合生成长度为 len(1)*len(2)*... 的索引 |
| .from_tuple() |
MI |
tuples ** |
[(第n行索引, )], tuple的第一个元素为一级标签, ... |
In[38]:
import pandas as pd
m_index1=pd.MultiIndex(levels=[['A','B'],['x1','x2']],labels=[[0,1],[0,1]],name=[None,"class2"])
m_index1
Out[38]:
MultiIndex(levels=[['A', 'B'], ['x1', 'x2']],
labels=[[0, 1], [0, 1]],
names=[None, 'class2'])
In [43]:
df1=pd.DataFrame(np.random.randint(1,10,(2,3)),index=m_index1)
df1
Out[43]:
| |
|
0 |
1 |
2 |
| |
class2 |
|
|
|
| A |
x1 |
1 |
4 |
9 |
| B |
x2 |
4 |
9 |
6 |
In [56]:
class1=['A','A','B','B']
class2=['x1','x2','y1','y2']
pd.MultiIndex.from_arrays([class1,class2],names=['class1','class2'])
Out[56]:
MultiIndex(levels=[['A', 'B'], ['x1', 'x2', 'y1', 'y2']],
labels=[[0, 0, 1, 1], [0, 1, 2, 3]],
names=['class1', 'class2'])
In [57]:
pd.MultiIndex.from_product([['A','B'],['x1','y1']],names=['class1','class2'])
Out[57]:
MultiIndex(levels=[['A', 'B'], ['x1', 'y1']],
labels=[[0, 0, 1, 1], [0, 1, 0, 1]],
names=['class1', 'class2'])