import numpy as np
import matplotlib.pyplot as plt # 与绘图有关的包都再pyplot下面
%matplotlib inline # 魔法函数，将绘图显示出来，否则绘图将存储在一个对象中
import pandas as pd
创建数组
np.random.randn(3,4)
array([[-0.76863424, 1.71936986, 0.10624708, 1.34709104],
[ 0.17115228, 2.03093016, -0.25014394, 1.44215241],
[ 0.49469013, 0.64961492, -0.12764838, 0.45850221]])
计算代码执行时间
%timeit [x**2 for x in range(4000)]
967 µs ± 102 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)
# Jupyter Notebook常用快捷键
# Tab键进行代码提示。查看方法，自动补全
# ？显示帮助文档。
# shift+Enter执行代码
# %matplotlib inline 将绘图直接显示出来
# %timeit 计算一段代码的执行时间
# a = np.array([True,True,False,False]) 创建bool类型是数组
# a.astype(np.float16) # 更改数据类型
a = np.array([[1,2,3],
[4,5,6]])
print(a.ndim)
print(a.shape)
print(a.dtype)
2
(2, 3)
int32
np.arange(36).reshape(3,3,4) # 3个3*4的数组
array([[[ 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]]])
np.linspace(0,1,100) # 从0-1 均匀生成100个数字。 常用于绘图
array([0. , 0.01010101, 0.02020202, 0.03030303, 0.04040404,
0.05050505, 0.06060606, 0.07070707, 0.08080808, 0.09090909,
0.1010101 , 0.11111111, 0.12121212, 0.13131313, 0.14141414,
0.15151515, 0.16161616, 0.17171717, 0.18181818, 0.19191919,
0.2020202 , 0.21212121, 0.22222222, 0.23232323, 0.24242424,
0.25252525, 0.26262626, 0.27272727, 0.28282828, 0.29292929,
0.3030303 , 0.31313131, 0.32323232, 0.33333333, 0.34343434,
0.35353535, 0.36363636, 0.37373737, 0.38383838, 0.39393939,
0.4040404 , 0.41414141, 0.42424242, 0.43434343, 0.44444444,
0.45454545, 0.46464646, 0.47474747, 0.48484848, 0.49494949,
0.50505051, 0.51515152, 0.52525253, 0.53535354, 0.54545455,
0.55555556, 0.56565657, 0.57575758, 0.58585859, 0.5959596 ,
0.60606061, 0.61616162, 0.62626263, 0.63636364, 0.64646465,
0.65656566, 0.66666667, 0.67676768, 0.68686869, 0.6969697 ,
0.70707071, 0.71717172, 0.72727273, 0.73737374, 0.74747475,
0.75757576, 0.76767677, 0.77777778, 0.78787879, 0.7979798 ,
0.80808081, 0.81818182, 0.82828283, 0.83838384, 0.84848485,
0.85858586, 0.86868687, 0.87878788, 0.88888889, 0.8989899 ,
0.90909091, 0.91919192, 0.92929293, 0.93939394, 0.94949495,
0.95959596, 0.96969697, 0.97979798, 0.98989899, 1. ])
a = np.arange(20) # 此时是浅拷贝，对拷贝出来的数据改变，会改变原始数组
# a[2:8:2] # 从第2位切分到第8位，每隔2位切分一个数字
# a[5]=100
# a
# b = a.copy() # 深拷贝
# b[0]=100 # 此时，对b的操作不会影响a数组
a > 10 # 生成一个bool值的array
array([False, False, False, False, False, False, False, False, False,
False, False, True, True, True, True, True, True, True,
True, True])
a[ a > 10 ] # 用bool值作为索引进行取值
array([11, 12, 13, 14, 15, 16, 17, 18, 19])
(a>5)&(a<15) # 获取bool数组
array([False, False, False, False, False, False, True, True, True,
True, True, True, True, True, True, False, False, False,
False, False])
a[(a>5)&(a<15)] # 将a位于5-15期间的数字索引出来
# a[(a>5)&(a<15)&(a%2==0)]
array([ 6, 7, 8, 9, 10, 11, 12, 13, 14])
a[(a<3)|(a>15)]
array([ 0, 1, 2, 16, 17, 18, 19])
b = np.arange(24).reshape(2,3,4)
b[(b>3)&(b%2!= 0)]
array([ 5, 7, 9, 11, 13, 15, 17, 19, 21, 23])
b[b<=10]
array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10])
b[~(b<=10)] # ~ 取反，非运算
array([11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23])
a = np.arange(10).reshape(2,5)
a
array([[0, 1, 2, 3, 4],
[5, 6, 7, 8, 9]])
广播
a+10 # 标量值广播，会广播到每一个元素.
# a*100
array([[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])
a+a # 相同位置对应元素相加
array([[ 0, 2, 4, 6, 8],
[10, 12, 14, 16, 18]])
a*a # 相同位置对应元素相乘
array([[ 0, 1, 4, 9, 16],
[25, 36, 49, 64, 81]])
a1 = np.arange(5)
a1
array([0, 1, 2, 3, 4])
a = np.arange(10).reshape(2,5)
a
array([[0, 1, 2, 3, 4],
[5, 6, 7, 8, 9]])
a1+a # a1与a只是列数相同，就可以进行加法运算。 # 列维度相同时，按照列广播
array([[ 0, 2, 4, 6, 8],
[ 5, 7, 9, 11, 13]])
b1 = np.arange(2).reshape(2,1)
b1
array([[0],
[1]])
a
array([[0, 1, 2, 3, 4],
[5, 6, 7, 8, 9]])
b1 + a # 行维度相同时，按照行广播
array([[ 0, 1, 2, 3, 4],
[ 6, 7, 8, 9, 10]])
a.T # 装置
array([[0, 5],
[1, 6],
[2, 7],
[3, 8],
[4, 9]])
a1 = np.array([[1,2],[3,4]])
b1 = np.array([[5,6],[7,8]])
np.concatenate((a1,b1))
array([[1, 2],
[3, 4],
[5, 6],
[7, 8]])
np.concatenate((a1,b1),axis=1)
array([[1, 2, 5, 6],
[3, 4, 7, 8]])
a1 = np.array([1,2,3,4])
a2 = np.array([5,6,7,8])
np.concatenate((a1,a2)) # 组合以后，还是一维数组
array([1, 2, 3, 4, 5, 6, 7, 8])
np.stack((a1,a2))# 组合以后，变成高维数组
array([[1, 2, 3, 4],
[5, 6, 7, 8]])
np.append(a1,5)
array([1, 2, 3, 4, 5])
b = np.stack((a1,a1))
b
array([[1, 2, 3, 4],
[1, 2, 3, 4]])
np.append(b,[9,10]) # append之后展开位1维数组
array([ 1, 2, 3, 4, 1, 2, 3, 4, 9, 10])
m = np.array([[1,2],[3,4]])
np.append(m,[5,6])
array([1, 2, 3, 4, 5, 6])
np.append(m,[[9,10]],axis=0)
array([[ 1, 2],
[ 3, 4],
[ 9, 10]])
np.append(m,[[9],[10]],axis=1) # 追加
array([[ 1, 2, 9],
[ 3, 4, 10]])
x = np.array([1,2,3,4,2,1,2,1,1,1,2,3,3,2,3])
np.unique(x) # 去重
array([1, 2, 3, 4])
y = np.arange(10)
np.square(y)
array([ 0, 1, 4, 9, 16, 25, 36, 49, 64, 81], dtype=int32)
y1 = np.arange(10,0,-1)
y1
array([10, 9, 8, 7, 6, 5, 4, 3, 2, 1])
np.maximum(y,y1) # 取出对应位置最大的那个数值
array([10, 9, 8, 7, 6, 5, 6, 7, 8, 9])
np.mean(y) # 顶级方法 顶级方法返回副本
# y.mean() # 对象方法
4.5
m
array([[1, 2],
[3, 4]])
m.mean(axis=1) # 按照行取均值
array([1.5, 3.5])
m.mean(axis=0) # 按照列取均值
array([2., 3.])
b
array([[1, 2, 3, 4],
[1, 2, 3, 4]])
b.max() # 取全局最大值
4
b.max(axis=0) # 取出每一列的最大值
array([1, 2, 3, 4])
b.max(axis=1) # 取出每一行的最大值
array([4, 4])
a = np.random.randn(19)
a
array([-1.61081374, -0.91845914, 0.76720491, 0.20492141, -1.15177974,
0.48665073, -0.68125193, 0.61895934, 0.87045498, 1.26682085,
-1.35425483, 0.33467291, 0.68378407, -0.21076582, -0.99360988,
0.11501658, -0.14558961, -1.13452916, -1.15467634])
a.sort()
a
# 如果用顶级方法，返回的是a排序后的副本，a本身不会发生变化
array([-1.61081374, -1.35425483, -1.15467634, -1.15177974, -1.13452916,
-0.99360988, -0.91845914, -0.68125193, -0.21076582, -0.14558961,
0.11501658, 0.20492141, 0.33467291, 0.48665073, 0.61895934,
0.68378407, 0.76720491, 0.87045498, 1.26682085])
b = np.random.randn(10)
b.shape=2,5
print("before sort:",b)
b.sort(axis=1) # 按照每一行进行排序
b
before sort: [[ 0.4619393 -0.30835792 0.55555801 0.1328978 0.13320337]
[-0.48944674 -1.50409789 -0.60189908 -1.19267168 -0.16516362]]
array([[-0.30835792, 0.1328978 , 0.13320337, 0.4619393 , 0.55555801],
[-1.50409789, -1.19267168, -0.60189908, -0.48944674, -0.16516362]])
c = np.array([2,4,1,3])
np.argsort(c)
# 排第一位，最小值是索引为2的值
# 排第2位 是索引为0的值
# 排第3位，是索引位3的值
# 排第4位，是索引为1的值
array([2, 0, 3, 1], dtype=int64)
np.argmax(c) # 返回最大值的索引
1
np.where(c>2) # 返回符合条件的索引
# 如果要找出这些值，直接用bool索引即可 c[c>2]
(array([1, 3], dtype=int64),)
d = np.array([9,5,6,7])
cond = np.array([True,False,False,False])
# 当condition为True时取c的值，False时取d的值
np.where(cond,c,d)
array([2, 5, 6, 7])
c
array([2, 4, 1, 3])
c>2
array([False, True, False, True])
(c>2).sum() #统计符合条件的元素的个数
2
np.random.randn(3,4)
array([[ 0.57739186, 0.56219424, -1.61210981, -0.74358949],
[ 0.34497086, 1.65096901, 0.14879807, 0.33655262],
[-0.09287764, 1.23421095, 0.36261984, -1.06369641]])
np.random.normal(size=(3,4))
array([[ 0.0284827 , -0.93254035, 0.41356179, 0.73310516],
[ 1.92833853, -1.13892997, 1.25694962, -1.31025191],
[ 0.13151099, -1.47289915, 0.45882324, 0.47906981]])
np.random.randint(1,10,(3,4))
array([[1, 2, 8, 4],
[8, 4, 6, 5],
[9, 8, 9, 9]])
c
array([2, 4, 1, 3])
np.random.permutation(len(c)) # 生成乱序的索引
array([0, 1, 2, 3])
c[np.random.permutation(len(c)) ] # 得到c的乱序的值
array([2, 1, 4, 3])
np.random.choice(c) # 从列表c中随机取出1个值
2
np.random.choice(c,size=2) # 从列表c中随机取出2个值
array([2, 2])
np.random.seed(5) # seed只生效一次
np.random.randn(3,4)
array([[ 0.44122749, -0.33087015, 2.43077119, -0.25209213],
[ 0.10960984, 1.58248112, -0.9092324 , -0.59163666],
[ 0.18760323, -0.32986996, -1.19276461, -0.20487651]])
np.random.seed(5) # 只要指定随机种子，每次生成的随机数相同。 伪随机
np.random.randn(3,4)
array([[ 0.44122749, -0.33087015, 2.43077119, -0.25209213],
[ 0.10960984, 1.58248112, -0.9092324 , -0.59163666],
[ 0.18760323, -0.32986996, -1.19276461, -0.20487651]])
np.save('variable_name',c) # 自动将c的值保存到variable_name文件中。默认文件后缀.npy
np.load('variable_name.npy')
array([2, 4, 1, 3])
np.savetxt('variable_name',c) # 以文本形式保存
np.loadtxt('variable_name')
array([2., 4., 1., 3.])
网友评论