0.背景

• stack将数据列旋转为行，unstack将数据的行旋转为列；
• pivot重塑的一个快捷方式。index，重塑的索引；columns，重塑的列名；values，生成新列的值；

1.stack & unstack

import numpy as np
import pandas as pd
data = pd.DataFrame(np.arange(4*5).reshape((4,5)),
                    index=pd.Index(['beijing','shanghai','guangzhou','shenzhen'],name='city'),
                    columns=pd.Index(['one','two','three','four','five'],name='number'))
print data

data

result_stack = data.stack(dropna=False)
print result_stack

result_stack.png

result_unstack = result_stack.unstack('city')
print result_unstack

result_unstack.png

2.pivot（目前在数据处理过程中，pivot不是必要的）

import numpy as np
import pandas as pd
data = pd.DataFrame(np.arange(4*5).reshape((4,5)),
                    index=pd.Index(['beijing','shanghai','guangzhou','shenzhen'],name='city'),
                    columns=pd.Index(['one','two','three','four','five'],name='number'))
result_pivot = data.pivot('one','two')
print result_pivot

result_pivot.png