pandas学习与使用1

学习了numpy之后，开始继续学习pandas。以下是pandas基本的一些语法及其使用方法，这一节主要是pandas中的Series结构。运行环境python2.7

#!/usr/bin/python
# -*- coding: UTF-8 -*-


import pandas as pd
import numpy as np

# Pandas模块的数据结构主要有两：1、Series ；2、DataFrame
# Series结构是基于NumPy的ndarray结构，是一个一维的标签矩阵
# pd.Series([list]，index=[list])
# 以list为参数，参数为一list; index为可选参数，若不填则默认index从0开始；若添则index长度与value长度相等

# 例子一
s = pd.Series([1, 2, 6, np.nan, 44, 1])
print "s:\n", s
# 例子二
s1 = pd.Series([1, 2, 3, 4, 5], index=["a", "b", "c", "d", "e"])
print "s1:\n", s1

# pd.Series({dict})以一字典结构为参数
s2 = pd.Series({'a': 3, 'b': 4, 'c': 5, 'd': 2, 'e': 6})
print "s2:\n", s2

# 取值 s[index] or s[[index的list]] 取值操作类似数组，当取不连续的多个值时可以以一list为参数
v = np.random.random_sample(20)
print "v:", v
l = pd.Series(v)
print "l:\n", l
l1 = l[[1, 5, 7, 9]]
print "l1:\n", l1
l2 = l[2:10]
print "l2:\n", l2
l3 = l[17]
print "l3:\n", l3

# head(n);tail(n)//取出头n行或尾n行，n为可选参数，若不填默认5
print "l的前五个值：\n", l.head()
print "l的后三个值：\n", l.tail(3)

# index(); values()//取出index 与values ,返回list
print type(l.index)   # class 'pandas.core.indexes.range.RangeIndex'
print "l的标签：\n", list(l.index)
print "l的值：\n", l.values

# Size、shape、uniqueness、counts of values
print "len():", len(l)  # Series长度,包括NaN    len(): 20
print "shape():", np.shape(l)  # 矩阵形状，（，）   shape(): (20L,)
print "count():", l.count()  # Series长度，不包括NaN      count(): 20
print "unique():", l.unique()  # 出现不重复values值
print "value_counts():\n", l.value_counts()  # 统计value值出现次数

# 加运算:相同index的value相加，若index并非共有的则该index对应value变为NaN

# s4: NaN 1 2 3 4
# s5: 1   1 1 1 NaN
# s6: NaN 2 3 4 NaN

s4 = pd.Series([1, 2, 3, 4], index=[1, 2, 3, 4])
s5 = pd.Series([1, 1, 1, 1])
s6 = s4 + s5
print "s4 + s5:\n", s6

以下是上述代码的运行结果

D:\software\Anaconda2\python.exe D:/PycharmProjects/Learn/learn_panda/learn_pd1.py
s:
0     1.0
1     2.0
2     6.0
3     NaN
4    44.0
5     1.0
dtype: float64
s1:
a    1
b    2
c    3
d    4
e    5
dtype: int64
s2:
a    3
b    4
c    5
d    2
e    6
dtype: int64
v: [0.37635755 0.4886929  0.07210501 0.79949712 0.45134537 0.87680035
 0.30748949 0.99859402 0.61613633 0.83520501 0.07343604 0.81827534
 0.25139748 0.65326514 0.45755433 0.31827714 0.3407211  0.07814553
 0.64210118 0.69003531]
l:
0     0.376358
1     0.488693
2     0.072105
3     0.799497
4     0.451345
5     0.876800
6     0.307489
7     0.998594
8     0.616136
9     0.835205
10    0.073436
11    0.818275
12    0.251397
13    0.653265
14    0.457554
15    0.318277
16    0.340721
17    0.078146
18    0.642101
19    0.690035
dtype: float64
l1:
1    0.488693
5    0.876800
7    0.998594
9    0.835205
dtype: float64
l2:
2    0.072105
3    0.799497
4    0.451345
5    0.876800
6    0.307489
7    0.998594
8    0.616136
9    0.835205
dtype: float64
l3:
0.0781455282331196
l的前五个值：
0    0.376358
1    0.488693
2    0.072105
3    0.799497
4    0.451345
dtype: float64
l的后三个值：
17    0.078146
18    0.642101
19    0.690035
dtype: float64
<class 'pandas.core.indexes.range.RangeIndex'>
l的标签：
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]
l的值：
[0.37635755 0.4886929  0.07210501 0.79949712 0.45134537 0.87680035
 0.30748949 0.99859402 0.61613633 0.83520501 0.07343604 0.81827534
 0.25139748 0.65326514 0.45755433 0.31827714 0.3407211  0.07814553
 0.64210118 0.69003531]
len(): 20
shape(): (20L,)
count(): 20
unique(): [0.37635755 0.4886929  0.07210501 0.79949712 0.45134537 0.87680035
 0.30748949 0.99859402 0.61613633 0.83520501 0.07343604 0.81827534
 0.25139748 0.65326514 0.45755433 0.31827714 0.3407211  0.07814553
 0.64210118 0.69003531]
value_counts():
0.451345    1
0.835205    1
0.307489    1
0.799497    1
0.818275    1
0.457554    1
0.616136    1
0.998594    1
0.078146    1
0.376358    1
0.318277    1
0.251397    1
0.073436    1
0.072105    1
0.488693    1
0.642101    1
0.690035    1
0.876800    1
0.340721    1
0.653265    1
dtype: int64
s4 + s5:
0    NaN
1    2.0
2    3.0
3    4.0
4    NaN
dtype: float64

Process finished with exit code 0