Numpy entry notes

As you know, we are in big data and artificial intelligence boom, python learning because of its low cost, flexible features became popular language. Numpy as a powerful python module, data analysis, machine learning and other situations have very important applications. Here follow the dog's head notes come together to learn about it!

Numpy installation

Method 1: download the official website to find the Getting Started you can find the installation method

Method 2: Use directly mounted pip

pip install numpy

Basics

First, open the python

1.import Numpy module

import numpy as np # 命名为np （约定俗成）

2. Matrix

Create a 2X3 matrix

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

Numerical matrix type definition

np.array([77,88,99], dtype=np.int32)
np.array([77,88,99], dtype=np.int64) # 默认的int
np.array([77,88,99], dtype=np.float32)
np.array([77,88,99], dtype=np.float64) # 默认的float

Create a 3X3 matrix elements are all 0 or 1

np.zeros((3,3))
np.ones((3,3))
np.ones((3,3),dtype=np.int16) # 同样的也可以改变他的数值类型

(Dummy) empty matrix

np.empty((2,2))

Generates an ordered sequence

np.arange(5) # [0 1 2 3 4]
np.arange(3,8) # [3 4 5 6 7]
np.arange(2,10,2) # [2 4 6 8]

You can also specify the number of elements in the sequence

np.linspace(1,10,5) # [1 3.25 5.5 7.75 10] 间隔9/4=2.25

By a sequence generator matrix

# 将[0,1,2,3,4,5,6,7,8] 变成 3X3 的矩阵（按行排列）
np.arange(9).reshape((3,3))

3. View Properties

Dimension of the matrix

matrix.ndim

Matrix shape

matrix.shape

The number of elements of the matrix

matrix.size

Numerical matrix of the type

matrix.type

Analyzing element size

matrix < 3 # 返回True 或 False 的列表
matrix > 3
matrix == 3

Summing element

np.sum(matrix)
np.sum(matrix, axis=0) # 求每一行的和
np.sum(matrix, axis=1) # 求每一列的和

Seeking maximum or minimum element

np.max(matrix)
np.max(matrix, axis=0) # 求每一行的最大值
np.max(matrix, axis=1) # 求每一列的最大值
np.min(matrix)
np.min(matrix, axis=0) # 求每一行的最小值
np.min(matrix, axis=1) # 求每一列的最小值

Seeking maximum or minimum value of the index

np.argmax(matrix)
np.argmin(matrix)
# 也可以指定行列，同上

Averaging, median, and accumulated, the accumulated difference

np.mean(matrix)
matrix.mean()
np.average(matrix)

np.median(matrix)

np.cumsum(matrix)
np.diff(matrix)

Progressive sorting

np.sort(matrix)

4. Simple operation

Addition and subtraction (addition and subtraction corresponding to the position of the element)

a = np.array([10, 20, 30, 40])
b = np.arange(4)
a + b
a - b

Multiplication and division (multiplication element corresponding to a position other)

a * b
a / b

Take the power of each element

b**2 # 平方
b**3 # 立方

Trigonometric functions

np.sin(a)
np.cos(a)
np.tan(a)

Matrix Multiplication

m1 = np.array([ [1,1], [0,1] ])
m2 = np.arange(4).reshape((2,2))
np.dot(m1,m2)
m1.dot(m2) # 两种方法一样

Matrix transpose

np.tanspose(m1)
m1.T

Cutting matrix

np.clip(matrix,3,8)
#小于3的元素都变成3，大于8的元素都变成8

The random number

Random generator matrix

np.random.seed(123)

np.random.random((2,4))
np.random.randint(8)
np.random.randint(4,10,size=6)

np.random.uniform() # 默认0到1的均匀分布
np.random.uniform(1,6)
numpy.random.normal(loc=0.0, scale=1.0, size=None)  # 正态分布随机

Continue to add. . . . . .