mirror of
https://github.com/apachecn/ailearning.git
synced 2026-04-23 18:13:05 +08:00
605 lines
22 KiB
Markdown
605 lines
22 KiB
Markdown
# Numpy 简介
|
||
|
||
## 导入numpy
|
||
|
||
**Numpy**是**Python**的一个很重要的第三方库，很多其他科学计算的第三方库都是以**Numpy**为基础建立的。
|
||
|
||
**Numpy**的一个重要特性是它的数组计算。
|
||
|
||
在使用**Numpy**之前，我们需要导入`numpy`包：
|
||
|
||
In [1]:
|
||
|
||
```py
|
||
from numpy import *
|
||
|
||
```
|
||
|
||
使用前一定要先导入 Numpy 包，导入的方法有以下几种：
|
||
|
||
```py
|
||
import numpy
|
||
import numpy as np
|
||
from numpy import *
|
||
from numpy import array, sin
|
||
|
||
```
|
||
|
||
事实上，在**ipython**中可以使用magic命令来快速导入**Numpy**的内容。
|
||
|
||
In [2]:
|
||
|
||
```py
|
||
%pylab
|
||
|
||
```
|
||
|
||
```py
|
||
Using matplotlib backend: Qt4Agg
|
||
Populating the interactive namespace from numpy and matplotlib
|
||
|
||
```
|
||
|
||
## 数组上的数学操作
|
||
|
||
假如我们想将列表中的每个元素增加`1`，但列表不支持这样的操作（报错）：
|
||
|
||
In [3]:
|
||
|
||
```py
|
||
a = [1, 2, 3, 4]
|
||
a + 1
|
||
|
||
```
|
||
|
||
```py
|
||
---------------------------------------------------------------------------
|
||
TypeError Traceback (most recent call last)
|
||
<ipython-input-3-068856d2a224> in <module>()
|
||
1 a = [1, 2, 3, 4]
|
||
----> 2 a + 1
|
||
|
||
TypeError: can only concatenate list (not "int") to list
|
||
```
|
||
|
||
转成 `array` ：
|
||
|
||
In [4]:
|
||
|
||
```py
|
||
a = array(a)
|
||
a
|
||
|
||
```
|
||
|
||
Out[4]:
|
||
|
||
```py
|
||
array([1, 2, 3, 4])
|
||
```
|
||
|
||
`array` 数组支持每个元素加 `1` 这样的操作：
|
||
|
||
In [5]:
|
||
|
||
```py
|
||
a + 1
|
||
|
||
```
|
||
|
||
Out[5]:
|
||
|
||
```py
|
||
array([2, 3, 4, 5])
|
||
```
|
||
|
||
与另一个 `array` 相加，得到对应元素相加的结果：
|
||
|
||
In [6]:
|
||
|
||
```py
|
||
b = array([2, 3, 4, 5])
|
||
a + b
|
||
|
||
```
|
||
|
||
Out[6]:
|
||
|
||
```py
|
||
array([3, 5, 7, 9])
|
||
```
|
||
|
||
对应元素相乘：
|
||
|
||
In [7]:
|
||
|
||
```py
|
||
a * b
|
||
|
||
```
|
||
|
||
Out[7]:
|
||
|
||
```py
|
||
array([ 2, 6, 12, 20])
|
||
```
|
||
|
||
对应元素乘方：
|
||
|
||
In [8]:
|
||
|
||
```py
|
||
a ** b
|
||
|
||
```
|
||
|
||
Out[8]:
|
||
|
||
```py
|
||
array([ 1, 8, 81, 1024])
|
||
```
|
||
|
||
## 提取数组中的元素
|
||
|
||
提取第一个元素：
|
||
|
||
In [9]:
|
||
|
||
```py
|
||
a[0]
|
||
|
||
```
|
||
|
||
Out[9]:
|
||
|
||
```py
|
||
1
|
||
```
|
||
|
||
提取前两个元素：
|
||
|
||
In [10]:
|
||
|
||
```py
|
||
a[:2]
|
||
|
||
```
|
||
|
||
Out[10]:
|
||
|
||
```py
|
||
array([1, 2])
|
||
```
|
||
|
||
最后两个元素：
|
||
|
||
In [11]:
|
||
|
||
```py
|
||
a[-2:]
|
||
|
||
```
|
||
|
||
Out[11]:
|
||
|
||
```py
|
||
array([3, 4])
|
||
```
|
||
|
||
将它们相加：
|
||
|
||
In [12]:
|
||
|
||
```py
|
||
a[:2] + a[-2:]
|
||
|
||
```
|
||
|
||
Out[12]:
|
||
|
||
```py
|
||
array([4, 6])
|
||
```
|
||
|
||
## 修改数组形状
|
||
|
||
查看 `array` 的形状：
|
||
|
||
In [13]:
|
||
|
||
```py
|
||
a.shape
|
||
|
||
```
|
||
|
||
Out[13]:
|
||
|
||
```py
|
||
(4L,)
|
||
```
|
||
|
||
修改 `array` 的形状：
|
||
|
||
In [14]:
|
||
|
||
```py
|
||
a.shape = 2,2
|
||
a
|
||
|
||
```
|
||
|
||
Out[14]:
|
||
|
||
```py
|
||
array([[1, 2],
|
||
[3, 4]])
|
||
```
|
||
|
||
## 多维数组
|
||
|
||
`a` 现在变成了一个二维的数组，可以进行加法：
|
||
|
||
In [15]:
|
||
|
||
```py
|
||
a + a
|
||
|
||
```
|
||
|
||
Out[15]:
|
||
|
||
```py
|
||
array([[2, 4],
|
||
[6, 8]])
|
||
```
|
||
|
||
乘法仍然是对应元素的乘积，并不是按照矩阵乘法来计算：
|
||
|
||
In [16]:
|
||
|
||
```py
|
||
a * a
|
||
|
||
```
|
||
|
||
Out[16]:
|
||
|
||
```py
|
||
array([[ 1, 4],
|
||
[ 9, 16]])
|
||
```
|
||
|
||
## 画图
|
||
|
||
linspace 用来生成一组等间隔的数据：
|
||
|
||
In [17]:
|
||
|
||
```py
|
||
a = linspace(0, 2*pi, 21)
|
||
%precision 3
|
||
a
|
||
|
||
```
|
||
|
||
Out[17]:
|
||
|
||
```py
|
||
array([ 0\. , 0.314, 0.628, 0.942, 1.257, 1.571, 1.885, 2.199,
|
||
2.513, 2.827, 3.142, 3.456, 3.77 , 4.084, 4.398, 4.712,
|
||
5.027, 5.341, 5.655, 5.969, 6.283])
|
||
```
|
||
|
||
三角函数：
|
||
|
||
In [18]:
|
||
|
||
```py
|
||
b = sin(a)
|
||
b
|
||
|
||
```
|
||
|
||
Out[18]:
|
||
|
||
```py
|
||
array([ 0.000e+00, 3.090e-01, 5.878e-01, 8.090e-01, 9.511e-01,
|
||
1.000e+00, 9.511e-01, 8.090e-01, 5.878e-01, 3.090e-01,
|
||
1.225e-16, -3.090e-01, -5.878e-01, -8.090e-01, -9.511e-01,
|
||
-1.000e+00, -9.511e-01, -8.090e-01, -5.878e-01, -3.090e-01,
|
||
-2.449e-16])
|
||
```
|
||
|
||
画出图像：
|
||
|
||
In [19]:
|
||
|
||
```py
|
||
%matplotlib inline
|
||
plot(a, b)
|
||
|
||
```
|
||
|
||
Out[19]:
|
||
|
||
```py
|
||
[<matplotlib.lines.Line2D at 0xa128ba8>]
|
||
```
|
||
|
||
|
||
|
||
## 从数组中选择元素
|
||
|
||
假设我们想选取数组b中所有非负的部分，首先可以利用 `b` 产生一组布尔值：
|
||
|
||
In [20]:
|
||
|
||
```py
|
||
b >= 0
|
||
|
||
```
|
||
|
||
Out[20]:
|
||
|
||
```py
|
||
array([ True, True, True, True, True, True, True, True, True,
|
||
True, True, False, False, False, False, False, False, False,
|
||
False, False, False], dtype=bool)
|
||
```
|
||
|
||
In [21]:
|
||
|
||
```py
|
||
mask = b >= 0
|
||
|
||
```
|
||
|
||
画出所有对应的非负值对应的点：
|
||
|
||
In [22]:
|
||
|
||
```py
|
||
plot(a[mask], b[mask], 'ro')
|
||
|
||
```
|
||
|
||
Out[22]:
|
||
|
||
```py
|
||
[<matplotlib.lines.Line2D at 0xa177be0>]
|
||
```
|
||
|
||
 |