import numpy as np
my_arr = np.arange(1000000)
my_list = list(range(1000000))
%time for _ in range(10):my_arr2 = my_arr * 2 #numpy
CPU times: user 27.2 ms, sys: 24.1 ms, total: 51.3 ms
Wall time: 67.1 ms
%time for _ in range(10):my_list2 = [x * 2 for x in my_list] #pure python
CPU times: user 892 ms, sys: 230 ms, total: 1.12 s
Wall time: 1.19 s
#产生随机数据
data = np.random.randn(2,3)
data
array([[ 1.31865052, 0.16184505, -0.19516902],
[-0.11928384, -1.53401869, 0.42106552]])
#乘法
data * 10
array([[ 13.18650519, 1.61845053, -1.95169017],
[ -1.1928384 , -15.34018694, 4.21065522]])
#加法
data + data
array([[ 2.63730104, 0.32369011, -0.39033803],
[-0.23856768, -3.06803739, 0.84213104]])
#shape,维度
data.shape
(2, 3)
#dtype,数据类型
data.dtype
dtype('float64')
#list to array
data1 = [6,7.5,8,0,1]
arr1 = np.array(data1)
arr1
array([6. , 7.5, 8. , 0. , 1. ])
#list to multidimensional array
data2 = [[1,2,3,4],[5,6,7,8]] #内嵌列表长度相等
arr2 = np.array(data2)
arr2
array([[1, 2, 3, 4],
[5, 6, 7, 8]])
#ndim 维度
arr2.ndim
2
arr2.shape #详细
(2, 4)
# 零
np.zeros(10)
np.zeros_like(10)
array(0)
# 多维零
np.zeros((3,6)) #括号数量
array([[0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0.]])
# 空
np.empty((2,3,2)) #产生的数字无意义
array([[[1., 0.],
[0., 0.],
[0., 1.]],
[[0., 0.],
[0., 0.],
[1., 0.]]])
# 生成1
np.ones(10)
array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1.])
#多维1
np.ones((3,6))
array([[1., 1., 1., 1., 1., 1.],
[1., 1., 1., 1., 1., 1.],
[1., 1., 1., 1., 1., 1.]])
np.eye(3) #矩阵
array([[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]])
np.eye(3,4) #行列不等
array([[1., 0., 0., 0.],
[0., 1., 0., 0.],
[0., 0., 1., 0.]])
np.identity(3) #矩阵
array([[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]])
# 生成序列,
np.arange(15)
array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14])
# asarray 并不复制
np.asarray(arr1)
array([6. , 7.5, 8. , 0. , 1. ])
