Python 清洗資料

標籤：資料   4.0   from   axis   建立   max   type   字典   class   

import numpy as npimport pandas as pdfrom pandas import Series,DataFrames=Series([1,2,3],index=[‘a‘,‘b‘,‘c‘])print(s)‘‘‘a    1b    2c    3dtype: int64‘‘‘print(np.max(s))#可以進行np運算s.name=‘rank‘s.index.name=‘name‘print(s)#建立DataFramesdata1={‘name‘:[‘a‘,‘b‘,‘c‘],‘rank‘:[1,2,3],‘score‘:[98,89,54]}print(sdata1)#字典df1=DataFrame(sdata1)print(df1)‘‘‘  name  rank  score0    a     1     981    b     2     892    c     3     54‘‘‘df2=DataFrame(sdata1,columns=[‘score‘,‘name‘,‘rank‘])print(df2)‘‘‘可以自動對齊，只是位置變化   score name  rank0     98    a     11     89    b     22     54    c     3‘‘‘df3=DataFrame(sdata1,columns=[‘score‘,‘name‘,‘rank‘,‘class‘],index=[‘1‘,‘2‘,‘3‘])print(df3)‘‘‘class這列是缺失值   score name  rank class1     98    a     1   NaN2     89    b     2   NaN3     54    c     3   NaN‘‘‘df4=df3.reindex([‘1‘,‘2‘,‘3‘,‘4‘])print(df4)‘‘‘重新索引   score name  rank class1   98.0    a   1.0   NaN2   89.0    b   2.0   NaN3   54.0    c   3.0   NaN4    NaN  NaN   NaN   NaN‘‘‘print(df4[‘score‘])print(df4.ix[‘1‘])print(df2[df2[‘score‘]>60])#返回df2中score大於60的值‘‘‘   score name  rank0     98    a     11     89    b     2‘‘‘del df3[‘class‘]print(df3)#刪除class這列sdata1={‘name‘:[‘a‘,‘b‘,‘c‘],‘rank‘:[1,2,3],‘score‘:[98,89,54]}print(sdata1)df3=DataFrame(sdata1,columns=[‘score‘,‘name‘,‘rank‘,‘class‘],index=[‘1‘,‘2‘,‘3‘])del df3[‘class‘]print(df3)print(df3.reindex([‘1‘,‘2‘,‘3‘,‘4‘]))print(df3.reindex([‘1‘,‘2‘,‘3‘,‘4‘],fill_value=0))#缺失值賦值為0‘‘‘   score name  rank1     98    a     12     89    b     23     54    c     34      0    0     0‘‘‘print(df3.reindex([‘0‘,‘1‘,‘2‘,‘3‘]))‘‘‘   score name  rank0    NaN  NaN   NaN1   98.0    a   1.02   89.0    b   2.03   54.0    c   3.0‘‘‘print(df3.reindex([‘0‘,‘1‘,‘2‘,‘3‘],method=‘bfill‘))#向後填充‘‘‘   score name  rank0     98    a     11     98    a     12     89    b     23     54    c     3‘‘‘print(df3.drop(‘1‘))#刪除第一行print(df3.drop(‘score‘,axis=1))#刪除指定列，axis是維數，0是行，1是列print(df3.T)#轉置

 

Python 清洗資料