机器学习入门(1)------python基础

环境：Python 3.6.4 |Anaconda, Inc.

Python常用容器类型

1.list

l = [1, 'a', 2, 'b']
print(type(l))
print('修改前：', l)

# 修改list的内容
l[0] = 3
print('修改后：', l)

# 末尾添加元素
l.append(4)
print('添加后：', l)

# 遍历list
print('遍历list(for循环)：')
for item in l:
    print(item)
    
# 通过索引遍历list
print('遍历list(while循环)：')
i = 0
while i != len(l):
    print(l[i])
    i += 1
    
# 列表合并
print('列表合并(+)：', [1, 2] + [3, 4])

# 列表重复
print('列表重复(*)：', [1, 2] * 5)

# 判断元素是否在列表中
print('判断元素存在(in)：', 1 in [1, 2])

<class 'list'>
修改前： [1, 'a', 2, 'b']
修改后： [3, 'a', 2, 'b']
添加后： [3, 'a', 2, 'b', 4]
遍历list(for循环)：
3
a
2
b
4
遍历list(while循环)：
3
a
2
b
4
列表合并(+)： [1, 2, 3, 4]
列表重复(*)： [1, 2, 1, 2, 1, 2, 1, 2, 1, 2]
判断元素存在(in)： True

2.tuple

t = (1, 'a', 2, 'b')
print(type(t))

#元组的内容不能修改，否则会报错
# t[0] = 3 

# 遍历tuple
print('遍历list(for循环)：')
for item in t:
    print(item)
    
# 通过索引遍历tuple
print('遍历tuple(while循环)：')
i = 0
while i != len(t):
    print(t[i])
    i += 1
    
# 解包 unpack
a, b, _, _ = t
print('unpack: ', c)

# 确保unpack接收的变量个数和tuple的长度相同，否则报错
# 经常出现在函数返回值的赋值时
# a, b, c = t

<class 'tuple'>
遍历list(for循环)：
1
a
2
b
遍历tuple(while循环)：
1
a
2
b

 

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-3-88506d8b1a51> in <module>()
     19 # 解包 unpack
     20 a, b, _, _ = t
---> 21 print('unpack: ', c)
     22 
     23 # 确保unpack接收的变量个数和tuple的长度相同，否则报错

NameError: name 'c' is not defined

3.dictionary

d = {'小象学院': 'http://www.chinahadoop.cn/',
    '百度': 'https://www.baidu.com/',
    '阿里巴巴': 'https://www.alibaba.com/',
    '腾讯': 'https://www.tencent.com/'}

print('通过key获取value: ', d['小象学院'])

# 遍历key
print('遍历key: ')
for key in d.keys():
    print(key)
    
# 遍历value
print('遍历value: ')
for value in d.values():
    print(value)
    
# 遍历item
print('遍历item: ')
for key, value in d.items():
    print(key + ': ' + value)

# format输出格式
print('format输出格式：')
for key, value in d.items():
    print('{}的网址是{}'.format(key, value))

通过key获取value:  http://www.chinahadoop.cn/
遍历key: 
小象学院
百度
阿里巴巴
腾讯
遍历value: 
http://www.chinahadoop.cn/
https://www.baidu.com/
https://www.alibaba.com/
https://www.tencent.com/
遍历item: 
小象学院: http://www.chinahadoop.cn/
百度: https://www.baidu.com/
阿里巴巴: https://www.alibaba.com/
腾讯: https://www.tencent.com/
format输出格式：
小象学院的网址是http://www.chinahadoop.cn/
百度的网址是https://www.baidu.com/
阿里巴巴的网址是https://www.alibaba.com/
腾讯的网址是https://www.tencent.com/

4.set

print('创建set:')
my_set = {1, 2, 3}
print(my_set)
my_set = set([1, 2, 3, 2])
print(my_set)

print('添加单个元素:')
my_set.add(3)
print('添加3', my_set)

my_set.add(4)
print('添加4', my_set)

print('添加多个元素：')
my_set.update([4, 5, 6])
print(my_set)

创建set:
{1, 2, 3}
{1, 2, 3}
添加单个元素:
添加3 {1, 2, 3}
添加4 {1, 2, 3, 4}
添加多个元素：
{1, 2, 3, 4, 5, 6}

5.Counter

• 初始化

import collections

c1 = collections.Counter(['a', 'b', 'c', 'a', 'b', 'b'])
c2 = collections.Counter({'a':2, 'b':3, 'c':1})
c3 = collections.Counter(a=2, b=3, c=1)

print(c1)
print(c2)
print(c3)

Counter({'b': 3, 'a': 2, 'c': 1})
Counter({'b': 3, 'a': 2, 'c': 1})
Counter({'b': 3, 'a': 2, 'c': 1})

• 更新内容

# 注意这里是做“加法”，不是“替换”
c1.update({'a': 4, 'c': -2, 'd': 4})
print(c1)

Counter({'a': 6, 'd': 4, 'b': 3, 'c': -1})

• 访问内容

print('a=', c1['a'])
print('b=', c1['b'])
# 对比和dict的区别
print('e=', c1['e'])

a= 6
b= 3
e= 0

• element()方法

for element in c1.elements():
    print(element)

d
d
d
d
b
b
b
a
a
a
a
a
a

• most_common()方法

c1.most_common(3)
[('a', 6), ('d', 4), ('b', 3)]

6.defaultdict

# 统计每个字母出现的次数
s = 'chinadoop'

# 使用Counter
print(collections.Counter(s))

Counter({'o': 2, 'd': 1, 'c': 1, 'p': 1, 'a': 1, 'n': 1, 'h': 1, 'i': 1})

# 使用dict
counter = {}
for c in s:
    if c not in counter:
        counter[c] = 1
    else:
        counter[c] += 1
        
print(counter.items())

dict_items([('d', 1), ('c', 1), ('p', 1), ('a', 1), ('o', 2), ('n', 1), ('h', 1), ('i', 1)])

# 使用defaultdict
counter2 = collections.defaultdict(int)
for c in s:
    counter2[c] += 1
print(counter2.items())

dict_items([('d', 1), ('c', 1), ('p', 1), ('a', 1), ('o', 2), ('n', 1), ('h', 1), ('i', 1)])

# 记录相同元素的列表
colors = [('yellow', 1), ('blue', 2), ('yellow', 3), ('blue', 4), ('red', 1)]
d = collections.defaultdict(list)
for k, v in colors:
    d[k].append(v)

print(d.items())

dict_items([('blue', [2, 4]), ('yellow', [1, 3]), ('red', [1])])

7.map函数

import math

print('示例1，获取两个列表对应位置上的最小值：')
l1 = [1, 3, 5, 7, 9]
l2 = [2, 4, 6, 6, 9]
mins = map(min, l1, l2)
print(mins)

# map()函数操作时，直到访问数据时才会执行
for item in mins:
    print(item)

print('示例2，对列表中的元素进行平方根操作：')
squared = map(math.sqrt, l2)
print(squared)
print(list(squared))

示例1，获取两个列表对应位置上的最小值：
<map object at 0x0000019AF8B0CDD8>
1
3
5
6
9
示例2，对列表中的元素进行平方根操作：
<map object at 0x0000019AF8A79DD8>
[1.4142135623730951, 2.0, 2.449489742783178, 2.449489742783178, 3.0]

8.匿名函数lambda

# my_func = lambda a, b, c: a * b
# print(my_func)
# print(my_func(1, 2, 3))

# 结合map
print('lambda结合map')
l1 = [1, 3, 5, 7, 9]
l2 = [2, 4, 6, 8, 10]
result = map(lambda x, y: x * 2 + y, l1, l2)
print(list(result))

lambda结合map
[4, 10, 16, 22, 28]

9.python操作csv数据文件

import csv

with open('grades.csv') as csvfile:
    grades_data = list(csv.DictReader(csvfile))
    
print('记录个数：', len(grades_data))
print('前2条记录：', grades_data[:2])
print('列名：', list(grades_data[0].keys()))

记录个数： 2315
前2条记录： [OrderedDict([('student_id', 'B73F2C11-70F0-E37D-8B10-1D20AFED50B1'), ('assignment1_grade', '92.73394640624123'), ('assignment1_submission', '2015-11-02 06:55:34.282000000'), ('assignment2_grade', '83.03055176561709'), ('assignment2_submission', '2015-11-09 02:22:58.938000000'), ('assignment3_grade', '67.16444141249367'), ('assignment3_submission', '2015-11-12 08:58:33.998000000'), ('assignment4_grade', '53.01155312999494'), ('assignment4_submission', '2015-11-16 01:21:24.663000000'), ('assignment5_grade', '47.710397816995446'), ('assignment5_submission', '2015-11-20 13:24:59.692000000'), ('assignment6_grade', '38.16831825359636'), ('assignment6_submission', '2015-11-22 18:31:15.934000000')]), OrderedDict([('student_id', '98A0FAE0-A19A-13D2-4BB5-CFBFD94031D1'), ('assignment1_grade', '86.79082085792986'), ('assignment1_submission', '2015-11-29 14:57:44.429000000'), ('assignment2_grade', '86.29082085792986'), ('assignment2_submission', '2015-12-06 17:41:18.449000000'), ('assignment3_grade', '69.7726566863439'), ('assignment3_submission', '2015-12-10 08:54:55.904000000'), ('assignment4_grade', '55.0981253490751'), ('assignment4_submission', '2015-12-13 17:32:30.941000000'), ('assignment5_grade', '49.5883128141676'), ('assignment5_submission', '2015-12-19 23:26:39.285000000'), ('assignment6_grade', '44.62948153275085'), ('assignment6_submission', '2015-12-21 17:07:24.275000000')])]
列名： ['student_id', 'assignment1_grade', 'assignment1_submission', 'assignment2_grade', 'assignment2_submission', 'assignment3_grade', 'assignment3_submission', 'assignment4_grade', 'assignment4_submission', 'assignment5_grade', 'assignment5_submission', 'assignment6_grade', 'assignment6_submission']

avg_assign1 = sum([float(row['assignment1_grade']) for row in grades_data]) / len(grades_data) 
print('assignment1平均分数：', avg_assign1)

assignment1平均分数： 74.5357320747794

assign1_sub_month = set(row['assignment1_submission'][:7] for row in grades_data)
print(assign1_sub_month)

{'2016-02', '2015-09', '2016-01', '2016-04', '2016-03', '2016-06', '2016-08', '2015-10', '2016-05', '2016-07', '2015-12', '2015-11'}

科学计算库NumPy

import numpy as np

1. 创建Array

my_list = [1, 2, 3]
x = np.array(my_list)

print('列表：', my_list)
print('Array: ', x)

列表： [1, 2, 3]
Array:  [1 2 3]

np.array([1, 2, 3]) - np.array([4, 5, 6])

array([-3, -3, -3])

m = np.array([[1, 2, 3], [4, 5, 6]])
print(m)
print('shape: ', m.shape)

[[1 2 3]
 [4 5 6]]
shape:  (2, 3)

n = np.arange(0, 30, 2)
print(n)

[ 0  2  4  6  8 10 12 14 16 18 20 22 24 26 28]

n = n.reshape(3, 5)
print('reshape后: ')
print(n)

reshape后: 
[[ 0  2  4  6  8]
 [10 12 14 16 18]
 [20 22 24 26 28]]

print('ones:
', np.ones((3, 2)))
print('zeros:
', np.zeros((3, 2)))
print('eye:
', np.eye(3))
print('diag:
', np.diag(my_list))

ones:
 [[1. 1.]
 [1. 1.]
 [1. 1.]]
zeros:
 [[0. 0.]
 [0. 0.]
 [0. 0.]]
eye:
 [[1. 0. 0.]
 [0. 1. 0.]
 [0. 0. 1.]]
diag:
 [[1 0 0]
 [0 2 0]
 [0 0 3]]

print('*操作：
', np.array([1, 2, 3] * 3))
print('repeat：
', np.repeat([1, 2, 3], 3))

*操作：
 [1 2 3 1 2 3 1 2 3]
repeat：
 [1 1 1 2 2 2 3 3 3]

p1 = np.ones((3, 3))
p2 = np.arange(9).reshape(3, 3)
print('纵向叠加: 
', np.vstack((p1, p2)))
print('横向叠加: 
', np.hstack((p1, p2)))

纵向叠加: 
 [[ 1.  1.  1.]
 [ 1.  1.  1.]
 [ 1.  1.  1.]
 [ 0.  1.  2.]
 [ 3.  4.  5.]
 [ 6.  7.  8.]]
横向叠加: 
 [[ 1.  1.  1.  0.  1.  2.]
 [ 1.  1.  1.  3.  4.  5.]
 [ 1.  1.  1.  6.  7.  8.]]

2. Array操作

p1 = np.array([[1, 1, 1], [1, 1, 1],[1,1,1]])
p2 = np.arange(9).reshape(3, 3)3 print('p1: 
', p1)
print('p2: 
', p2)

print('p1 + p2 = 
', p1 + p2)
print('p1 * p2 = 
', p1 * p2)
print('p2^2 = 
', p2 ** 2)
print('p1.p2 = 
', p1.dot(p2))

p1: 
 [[1 1 1]
 [1 1 1]
 [1 1 1]]
p2: 
 [[0 1 2]
 [3 4 5]
 [6 7 8]]
p1 + p2 = 
 [[1 2 3]
 [4 5 6]
 [7 8 9]]
p1 * p2 = 
 [[0 1 2]
 [3 4 5]
 [6 7 8]]
p2^2 = 
 [[ 0  1  4]
 [ 9 16 25]
 [36 49 64]]
p1.p2 = 
 [[ 9 12 15]
 [ 9 12 15]
 [ 9 12 15]]

p3 = np.arange(6).reshape(2, 3)
print('p3形状: ', p3.shape)
print(p3)
p4 = p3.T
print('转置后p3形状: ', p4.shape)
print(p4)

p3形状:  (2, 3)
[[0 1 2]
 [3 4 5]]
转置后p3形状:  (3, 2)
[[0 3]
 [1 4]
 [2 5]]

p3 = np.arange(6).reshape(2, 3)
print('p3数据类型:', p3.dtype)
print(p3)

p5 = p3.astype('float')
print('p5数据类型:', p5.dtype)
print(p5)

p3数据类型: int32
[[0 1 2]
 [3 4 5]]
p5数据类型: float64
[[0. 1. 2.]
 [3. 4. 5.]]

a = np.array([-4, -2, 1, 3, 5])
print('sum: ', a.sum())
print('min: ', a.min())
print('max: ', a.max())
print('mean: ', a.mean())
print('std: ', a.std()) //标准差
print('argmax: ', a.argmax()) //argmax(f(x))是使得 f(x)取得最大值所对应的变量x
print('argmin: ', a.argmin()) //argmax(f(x))是使得 f(x)取得最小值所对应的变量x

sum:  3
min:  -4
max:  5
mean:  0.6
std:  3.2619012860600183
argmax:  4
argmin:  0

3. 索引与切片

# 一维array
s = np.arange(13) ** 2
print('s: ', s)
print('s[0]: ', s[0])
print('s[4]: ', s[4])
print('s[0:3]: ', s[0:3])
print('s[[0, 2, 4]]: ', s[[0, 2, 4]])

s:  [  0   1   4   9  16  25  36  49  64  81 100 121 144]
s[0]:  0
s[4]:  16
s[0:3]:  [0 1 4]
s[[0, 2, 4]]:  [ 0  4 16]

# 二维array
r = np.arange(36).reshape((6, 6))
print('r: 
', r)
print('r[2, 2]: 
', r[2, 2]) //对应矩阵第三行第三列
print('r[3, 3:6]: 
', r[3, 3:6]) //对应第四行第四列到第7列的数（只表示该行的数）

r: 
 [[ 0  1  2  3  4  5]
 [ 6  7  8  9 10 11]
 [12 13 14 15 16 17]
 [18 19 20 21 22 23]
 [24 25 26 27 28 29]
 [30 31 32 33 34 35]]
r[2, 2]: 
 14
r[3, 3:6]: 
 [21 22 23]

r = np.arange(36).reshape((6, 6))
r > 30

array([[False, False, False, False, False, False],
       [False, False, False, False, False, False],
       [False, False, False, False, False, False],
       [False, False, False, False, False, False],
       [False, False, False, False, False, False],
       [False,  True,  True,  True,  True,  True]])

# 过滤
print(r[r > 30])

# 将大于30的数赋值为30
r[r > 30] = 30
print(r)

[31 32 33 34 35]
[[ 0  1  2  3  4  5]
 [ 6  7  8  9 10 11]
 [12 13 14 15 16 17]
 [18 19 20 21 22 23]
 [24 25 26 27 28 29]
 [30 30 30 30 30 30]]

# copy()操作
r2 = r[:3, :3]
print(r2)

[[ 0  1  2]
 [ 6  7  8]
 [12 13 14]]

# 将r2内容设置为0
r2[:] = 0

# 查看r的内容
print(r)

[[ 0  0  0  3  4  5]
 [ 0  0  0  9 10 11]
 [ 0  0  0 15 16 17]
 [18 19 20 21 22 23]
 [24 25 26 27 28 29]
 [30 30 30 30 30 30]]

r3 = r.copy()
r3[:] = 0
print(r)

[[ 0  0  0  3  4  5]
 [ 0  0  0  9 10 11]
 [ 0  0  0 15 16 17]
 [18 19 20 21 22 23]
 [24 25 26 27 28 29]
 [30 30 30 30 30 30]]

4. 遍历 Array

import numpy as np
t = np.random.randint(0, 10, (4, 3))
print(t)

[[3 2 7]
 [4 9 1]
 [1 3 0]
 [0 9 1]]

for row in t:
    print(row)

[3 2 7]
[4 9 1]
[1 3 0]
[0 9 1]

# 使用enumerate()
for i, row in enumerate(t):
    print('row {} is {}'.format(i, row))

row 0 is [3 2 7]
row 1 is [4 9 1]
row 2 is [1 3 0]
row 3 is [0 9 1]

t2 = t ** 2
print(t2)

[[ 9  4 49]
 [16 81  1]
 [ 1  9  0]
 [ 0 81  1]]

# 使用zip对两个array进行遍历计算
for i, j in zip(t, t2):
    print('{} + {} = {}'.format(i, j, i + j))

[3 2 7] + [ 9  4 49] = [12  6 56]
[4 9 1] + [16 81  1] = [20 90  2]
[1 3 0] + [1 9 0] = [ 2 12  0]
[0 9 1] + [ 0 81  1] = [ 0 90  2]