python学习----pandas

1.Series

s1=pd.Series([1,2,3,4])
print(s1)
#输出数据-------
#0    1
# 1    2
# 2    3
# 3    4
# dtype: int64

#改变他的索引
s2=pd.Series([1,2,3,4],index=("a","b","c","d"))
print(s2)
# a    1
# b    2
# c    3
# d    4
# dtype: int64

#查询
s2["a"]=100
print(
    s2
)

#属性打印
print(s2.index)  #索引
print(s2.values) #值
print(s2.shape) #4行
print(s2.ndim)  #
print(s2.size)
print(s2.dtype)
print(s2.head())  #钱几行,默认是前五行
s2.index.name="ok"
print(s2)
print(s2.index.name)
# ok
# a    100
# b      2
# c      3
# d      4
# dtype: int64
# ok

2.DataFrame

#通过pd.DateFrame 创建
df1=pd.DataFrame([[1,2,3],[4,5,6],[7,8,9]])
print(df1)
#    0  1  2
# 0  1  2  3
# 1  4  5  6
# 2  7  8  9

#修改行,列索引名
df2=pd.DataFrame([[1,2,3],[4,5,6],[7,8,9]],index=("a","b","c"),columns=("A","B","C"))
print(df2)
#    A  B  C
# a  1  2  3
# b  4  5  6
# c  7  8  9

#通过numpy 创建
import numpy as np
df3=pd.DataFrame(np.random.randn(3,3),index=("a","b","c"),columns=("A","B","C"))
print(df3)
#           A         B         C
# # a -1.532409 -2.349603  0.744708
# # b -1.425400 -0.706411 -1.384674
# # c  1.025274 -0.933366  0.188983
print("--------打印行或列的信息-----")
print(df3["A"])
#DataFrame 中的ix被废弃  loc是通过通过行标签索引行数据  iloc——通过行号索引行数据

print(df2)
#    A  B  C
# a  1  2  3
# b  4  5  6
# c  7  8  9

print(df2.loc["a","A"])
# 1

print(df2.iloc[0,1])
# 2


print(df2.loc[:,"A"])
# a    1
# b    4
# c    7
# Name: A, dtype: int64

print(df2.loc["a",:])
#A    1
# B    2
# C    3
# Name: a, dtype: int64

#更改 数据
df2.loc["a","A"]=100

#删除 数据   --只能删除列
del df2["A"]
print(df2)
#    B  C
# a  2  3
# b  5  6
# c  8  9

#增加
df2["D"]=[1,2,3]

3对齐运算

print("----------对其运算---------------")
import pandas as pd

s1=pd.Series(data=range(5),index=range(5))
s2=pd.Series(data=range(10),index=range(10))
print(s1+s2)
# 0    0.0
# 1    2.0
# 2    4.0
# 3    6.0
# 4    8.0
# 5    NaN
# 6    NaN
# 7    NaN
# 8    NaN
# 9    NaN
#对齐运算
print(s1.add(s2,fill_value=100))
# 0      0.0
# 1      2.0
# 2      4.0
# 3      6.0
# 4      8.0
# 5    105.0
# 6    106.0
# 7    107.0
# 8    108.0
# 9    109.0

4.pandas的函数应用

import pandas as pd
import numpy as np

df1 = pd.DataFrame([np.random.randn(3), [np.nan, 2, 2], [1, np.nan, 3]])
print(df1)
#           0         1         2
# 0 -0.307309 -0.602356  0.468423
# 1       NaN  2.000000  2.000000
# 2  1.000000       NaN  3.000000
#isnull 判断是否为空
print(df1.isnull())
#        0      1      2
# 0  False  False  False
# 1   True  False  False
# 2  False   True  False

#fillna  填充空值
print(df1.fillna(100))
#            0           1         2
# 0   -1.61506    0.472201  0.518066
# 1  100.00000    2.000000  2.000000
# 2    1.00000  100.000000  3.000000

 #dropna#
#默认删除行
print(df1.drop([0]))
#      0    1    2
# 1  NaN  2.0  2.0
# 2  1.0  NaN  3.0

#删除列
print(df1.drop(columns=1))
#           0         2
# 0 -0.630894  0.385834
# 1       NaN  2.000000
# 2  1.000000  3.000000

5层次索引

import numpy as np
import pandas as pd
s1=pd.Series(np.random.randn((12)),
          index=[["a","a","a","b","b","b","c","c","c","d","d","d",],
                 ["1","2","3","1","2","3","1","2","3","1","2","3"]])
print(s1)
# a  1    0.584870
# #    2   -0.137987
# #    3   -1.070576
# # b  1   -0.519936
# #    2   -1.796467
# #    3    1.153866
# # c  1   -0.450462
# #    2    1.204522
# #    3    0.726738
# # d  1    1.219730
# #    2   -0.364476
# #    3    1.489353

print(s1["a"][1])
# -0.03686968088802105

s1["a"][1]=100
print(s1["a"][1])
# 100.0

#index互换
print(s1.swaplevel())
# 1  a      1.861300
# 2  a    100.000000
# 3  a      1.151022
# 1  b     -0.003058
# 2  b     -0.781960
# 3  b     -1.819286
# 1  c     -0.339151
# 2  c     -1.028886
# 3  c      0.099268
# 1  d     -0.262582
# 2  d     -0.308409
# 3  d     -0.198745
# dtype: float64

#根据索引排序
print(s1.swaplevel().sort_index())
# 1  a      1.861300
#    b     -0.003058
#    c     -0.339151
#    d     -0.262582
# 2  a    100.000000
#    b     -0.781960
#    c     -1.028886
#    d     -0.308409
# 3  a      1.151022
#    b     -1.819286
#    c      0.099268
#    d     -0.198745
# dtype: float64

6.pandas统计计算

import numpy as np
import  pandas as pd

s1 = pd.DataFrame(np.random.randn(5, 4), index=("A", "B", "C", "D", "E"), columns=("a", "b", "c", "d"))
print(s1)

#计算函数 sum  mean  max  min
#按行统计  axis = 1
#按列统计   axis=0
# skipna 排除缺失值  ,默认为True
print(s1.min(axis=0))
#           a         b         c         d
# A  0.256362 -0.821711  1.404762  0.236821
# B  0.368196  0.466041 -0.080604  0.561946
# C -0.497315  0.094333  2.598284  0.358305
# D  0.588792 -0.297474 -1.556710  0.006757
# E  1.396045 -0.214725  0.397240 -1.023083
# a   -0.497315
# b   -0.821711
# c   -1.556710
# d   -1.023083
# dtype: float64

print(s1.min(axis=1))
# A   -0.683147
# B   -1.239646
# C   -1.699261
# D   -1.294243
# E   -1.004114
# dtype: float64

print(s1.describe())
#               a         b         c         d
# count  5.000000  5.000000  5.000000  5.000000   有几个样本
# mean  -0.577773 -0.205601 -0.068531 -0.192949     均值
# std    1.162071  0.367125  0.979809  0.956035     标准差
# min   -2.410131 -0.689078 -1.225277 -1.301825     最小值
# 25%   -0.915416 -0.300361 -0.534038 -0.785917     1/4分位数
# 50%   -0.302638 -0.225946 -0.450630 -0.386991     中位数
# 75%    0.197694 -0.148005  0.678076  0.408620     3/4分位数
# max    0.541625  0.335383  1.189214  1.101367     最大值

#query 根据给的条件选出数据(只有列?)
print(s1)
print(s1.query("a>b"))
#           a         b         c         d
# A -1.641773 -1.350308 -0.597922  0.195105
# B -0.003405  0.551447 -0.400094 -2.092777
# C -0.980596 -1.048407 -1.933153  0.610606
# D  0.672218  1.151102 -0.810194  0.921114
# E -0.172449  0.199802 -0.427191 -0.062688
#           a         b         c         d
# C -0.980596 -1.048407 -1.933153  0.610606

8.pandas读取文件

import pandas as pd
file=pd.read_csv("./SklearnTest.txt")
print(file)

#取出一列
print(file["height"])
print(file.height)

#取出两列
print(file.loc[:,"height":"house"])
print(file.iloc[:,0:2])
#    height  house
# 0    1.80      1
# 1    1.62      1
# 2    1.71      0
# 3    1.58      1
# 4    1.68      0
# 5    1.63      1
# 6    1.78      0
# 7    1.64      0
# 8    1.65      0

#取出样本数据0-5
print(file.iloc[0:6,:])
print(file.loc[0:6,:])
#    height  house  car  handsome  job  is_date
# 0    1.80      1    0       6.5    2        1
# 1    1.62      1    0       5.5    0        1
# 2    1.71      0    1       8.5    1        1
# 3    1.58      1    1       6.3    1        1
# 4    1.68      0    1       5.1    0        0
# 5    1.63      1    0       5.3    1        0
# 6    1.78      0    0       4.5    0        0

#选择特征列和列别标签列
X1=file.loc[:,"height":"job"]
print(X1)
print(type(X1))
#    height  house  car  handsome  job
# # 0    1.80      1    0       6.5    2
# # 1    1.62      1    0       5.5    0
# # 2    1.71      0    1       8.5    1
# # 3    1.58      1    1       6.3    1
# # 4    1.68      0    1       5.1    0
# # 5    1.63      1    0       5.3    1
# # 6    1.78      0    0       4.5    0
# # 7    1.64      0    0       7.8    2
# # 8    1.65      0    1       6.6    0
# # <class 'pandas.core.frame.DataFrame'>

Y1 = file.loc[:,"is_date"]
print(Y1)
print(type(Y1))
# 0    1
# 1    1
# 2    1
# 3    1
# 4    0
# 5    0
# 6    0
# 7    0
# 8   -1
# Name: is_date, dtype: int64
# <class 'pandas.core.series.Series'>

#选择is_date=
new_date = file.query("is_date!=-1")
date = file.query("is_date==-1")
print(new_date)
#    height  house  car  handsome  job  is_date
# 0    1.80      1    0       6.5    2        1
# 1    1.62      1    0       5.5    0        1
# 2    1.71      0    1       8.5    1        1
# 3    1.58      1    1       6.3    1        1
# 4    1.68      0    1       5.1    0        0
# 5    1.63      1    0       5.3    1        0
# 6    1.78      0    0       4.5    0        0
# 7    1.64      0    0       7.8    2        0
print(date)
#    height  house  car  handsome  job  is_date
# 8    1.65      0    1       6.6    0       -1

#将处理好的文件分为X Y
X = new_date.iloc[:,0:5]
Y = new_date.iloc[:,5]
print("------------------------")
print(X)
print(type(X))
#    height  house  car  handsome  job
# 0    1.80      1    0       6.5    2
# 1    1.62      1    0       5.5    0
# 2    1.71      0    1       8.5    1
# 3    1.58      1    1       6.3    1
# 4    1.68      0    1       5.1    0
# 5    1.63      1    0       5.3    1
# 6    1.78      0    0       4.5    0
# 7    1.64      0    0       7.8    2
# <class 'pandas.core.frame.DataFrame'>

print("-----------------------")
print(Y)
print(type(Y))
# 0    1
# 1    1
# 2    1
# 3    1
# 4    0
# 5    0
# 6    0
# 7    0
# Name: is_date, dtype: int64
# <class 'pandas.core.series.Series'>

9.pandas函数补充

9.1 cocat
pandas.concat()通常用来连接DataFrame对象。默认情况下是对两个DataFrame对象进行纵向连接， 当然通过设置参数，也可以通过它实现DataFrame对象的横向连接(axis=1)

import pandas as pd
s1= pd.Series(['a','b'])
s2= pd.Series(['c','d'])
print(s1)
# 0    a
# 1    b

print(s2)
# 0    c
# 1    d

print(pd.concat([s1,s2],axis=0))
# 0    a
# 1    b
# 0    c
# 1    d

print(pd.concat([s1,s2],axis=1))
#    0  1
# 0  a  c
# 1  b  d


#如果希望重新设置合并之后的DataFrame对象的index值， 可以添加ignore_index=True参数：
df1 = pd.DataFrame([['a', 1], ['b', 2]], columns=['letter', 'number'])
print(df1)
  # letter  number
# 0      a       1
# 1      b       2

df2 = pd.DataFrame([['c', 3], ['d', 4]], columns=['letter', 'number'])
print(df2)
#   letter  number
# 0      c       3
# 1      d       4

print(pd.concat([df1,df2] ,ignore_index=True))
#   letter  number
# 0      a       1
# 1      b       2
# 2      c       3
# 3      d       4

print(pd.concat([df1,df2] ,ignore_index=True,axis=1))
#    0  1  2  3
# 0  a  1  c  3
# 1  b  2  d  4

2.groupby函数

import pandas as pd
s1 = pd.DataFrame([[1,2,3],[1,3,3],[4,5,6],[4,3,6]],columns=["one","two","three"])
print(s1)
#    one  two  three
# 0    1    2      3
# 1    1    3      3
# 2    4    5      6
# 3    4    3      6
print("--------------------")

print(s1.groupby(by = ["one"])["two"])
# <pandas.core.groupby.generic.SeriesGroupBy object at 0x0000026484B259E8>
print("0--------------------")
#先分组根据 "one"  然后再选区"two" 的平均值
print(s1.groupby(by = ["one"])["two"].mean())
# one
# 1    2.5 ->根据均值来选的  (2+3)/2=2.5
# 4    4.0

#等价写法
print(s1["two"].groupby(by=s1["one"]).mean())
# one
# 1    2.5
# 4    4.0
# Name: two, dtype: float64


import numpy as np
#根据one进行聚合,再对其余列进行求解均值
print(s1.groupby(["one"]).mean())
#      two  three
# one
# 1    2.5    3.0
# 4    4.0    6.0

#根据one列进行聚合,再对其余列进行求解均值
print(s1.groupby(["one"]).transform(lambda  x:np.mean(x)))
#    two  three
# 0  2.5    3.0
# 1  2.5    3.0
# 2  4.0    6.0
# 3  4.0    6.0

print(s1.groupby(["one"]).transform(lambda  x:(x-np.mean(x))/(np.std(x))))
#    two  three
# 0 -1.0    NaN
# 1  1.0    NaN
# 2  1.0    NaN
# 3 -1.0    NaN

#根据one列进行聚合,再对制定列进行求解均值,
print(s1.groupby(["one"]).apply(lambda  s1:np.mean(s1["two"])))
# one
# 1    2.5
# 4    4.0
# dtype: float64

检查缺失值