# NumPy基础
#NumPy的ndarray：一种多维数组对象
import numpy as np
data = np.random.randn(2,3)
data
array([[ 0.93326604, 1.25437496, 0.55385005],
[ 1.54395985, 1.77476317, -0.83043758]])
data*10
array([[-24.31253818, 10.96863302, -3.7253224 ],
[ 6.94385942, 1.56600743, 0.6273361 ]])
data+data
array([[-4.86250764, 2.1937266 , -0.74506448],
[ 1.38877188, 0.31320149, 0.12546722]])
#ndarray:一个通用的同构数据多维容器，元素相同类型
#每个数组都有一个shape（表示各维度大小的元组）和一个dtype（说明数组数据类型的对象）
data.shape
(2, 3)
data.dtype
dtype('float64')
#创建ndarray数组
data1 = [6,7.5,8,0,1]
arr1 = np.array(data1)
arr1
array([6. , 7.5, 8. , 0. , 1. ])
data2 = [[1, 2, 3, 4], [5, 6, 7, 8]]
arr2 = np.array(data2)
arr2
array([[1, 2, 3, 4],
[5, 6, 7, 8]])
arr2.shape
(2, 4)
arr2.dtype
dtype('int32')
np.zeros(10)
array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])
np.zeros((3,6))
array([[0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0.]])
np.empty((2,3,2)) #未初始化的垃圾值
array([[[1.25712451e-311, 2.47032823e-322],
[0.00000000e+000, 0.00000000e+000],
[0.00000000e+000, 5.64233733e-067]],
[[4.76161762e-090, 5.64053486e-062],
[2.32597619e-057, 2.06031264e+184],
[3.99910963e+252, 2.59194823e+184]]])
np.ones(10)
array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1.])
np.ones((2,3))
array([[1., 1., 1.],
[1., 1., 1.]])
np.arange(15)
array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14])
#NumPy数组的运算
arr = np.array([[1,2,3],[4,5,6]])
arr
array([[1., 2., 3.],
[4., 5., 6.]])
arr*arr
array([[ 1., 4., 9.],
[16., 25., 36.]])
arr-arr
array([[0., 0., 0.],
[0., 0., 0.]])
1/arr
array([[1. , 0.5 , 0.33333333],
[0.25 , 0.2 , 0.16666667]])
arr**0.5
array([[1. , 1.41421356, 1.73205081],
[2. , 2.23606798, 2.44948974]])
arr2 = np.array([[0,4,1],[7,2,12]])
arr2
array([[ 0, 4, 1],
[ 7, 2, 12]])
arr2>arr
array([[False, True, False],
[ True, False, True]])
#基本的索引和切片
arr = np.arange(10)
arr
array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
arr[5]
5
arr[6:8]
array([6, 7])
arr[5:8] = 12
arr
array([ 0, 1, 2, 3, 4, 12, 12, 12, 8, 9])
arr_slice = arr[5:8]
arr_slice
array([12, 12, 12])
arr_slice[1] = 123
arr
array([ 0, 1, 2, 3, 4, 12, 123, 12, 8, 9])
arr_slice[:]=0
arr
array([0, 1, 2, 3, 4, 0, 0, 0, 8, 9])
arr2d = np.array([[1,2,3],[4,5,6],[7,8,9]])
arr2d[2]
array([7, 8, 9])
arr2d[2][1]
8
arr2d[2,1]
8
arr3d = np.array([[[1, 2, 3], [4, 5, 6]], [[7, 8, 9], [10, 11, 12]]])#2*2*3
arr3d
array([[[ 1, 2, 3],
[ 4, 5, 6]],
[[ 7, 8, 9],
[10, 11, 12]]])
arr3d[0]
array([[1, 2, 3],
[4, 5, 6]])
old_values = arr3d[0].copy()
arr3d[0] =42
arr3d
array([[[42, 42, 42],
[42, 42, 42]],
[[ 7, 8, 9],
[10, 11, 12]]])
arr3d[0] = old_values
arr3d
array([[[ 1, 2, 3],
[ 4, 5, 6]],
[[ 7, 8, 9],
[10, 11, 12]]])
arr3d[1,0]
array([7, 8, 9])
x = arr3d[1]
x
array([[ 7, 8, 9],
[10, 11, 12]])
x[0]
array([7, 8, 9])
#切片索引
arr
array([0, 1, 2, 3, 4, 0, 0, 0, 8, 9])
arr[1:6]
array([1, 2, 3, 4, 0])
arr2d
array([[1, 2, 3],
[4, 5, 6],
[7, 8, 9]])
arr2d[:2]
array([[1, 2, 3],
[4, 5, 6]])
arr2d[:2,1:]#切片不降维
array([[2, 3],
[5, 6]])
arr2d[1,:2]#切片+索引
array([4, 5])
arr2d[:2,2]
array([3, 6])
arr2d[2,:2]
array([7, 8])
arr2d[:,:1]
array([[1],
[4],
[7]])
arr2d[:2,1:] = 0
arr2d
array([[1, 0, 0],
[4, 0, 0],
[7, 8, 9]])
#布尔型索引
names = np.array(['Bob', 'Joe', 'Will', 'Bob', 'Will', 'Joe', 'Joe'])
data = np.random.randn(7,4)
names
array(['Bob', 'Joe', 'Will', 'Bob', 'Will', 'Joe', 'Joe'], dtype='<U4')
data
array([[-0.76003934, 1.20343331, -0.68892317, 0.43826464],
[-1.02005597, 0.74198114, -0.23101841, -0.67845112],
[-0.19384601, 3.36221251, 0.23700581, 0.16210278],
[-0.22421591, -0.04528472, 0.83725855, -1.01209329],
[ 0.13628598, 0.35140916, 1.05812391, 0.27382559],
[ 1.8598765 , 1.25615814, 0.8924981 , 0.6721845 ],
[-0.70007119, 0.17505724, -0.76280317, 2.35889359]])
names == 'Bob'
array([ True, False, False, True, False, False, False])
data[names == 'Bob']
array([[-0.76003934, 1.20343331, -0.68892317, 0.43826464],
[-0.22421591, -0.04528472, 0.83725855, -1.01209329]])
data[names == 'Bob',2:]
array([[-0.68892317, 0.43826464],
[ 0.83725855, -1.01209329]])
data[names == 'Bob',3]
array([ 0.43826464, -1.01209329])
names!= 'Bob'
array([False, True, True, False, True, True, True])
data[~(names == 'Bob')]
array([[-1.02005597, 0.74198114, -0.23101841, -0.67845112],
[-0.19384601, 3.36221251, 0.23700581, 0.16210278],
[ 0.13628598, 0.35140916, 1.05812391, 0.27382559],
[ 1.8598765 , 1.25615814, 0.8924981 , 0.6721845 ],
[-0.70007119, 0.17505724, -0.76280317, 2.35889359]])
mask = (names == 'Bob') |(names == 'Will')
mask
array([ True, False, True, True, True, False, False])
data[mask]
array([[-0.76003934, 1.20343331, -0.68892317, 0.43826464],
[-0.19384601, 3.36221251, 0.23700581, 0.16210278],
[-0.22421591, -0.04528472, 0.83725855, -1.01209329],
[ 0.13628598, 0.35140916, 1.05812391, 0.27382559]])
data[data<0] = 0
data
array([[0. , 1.20343331, 0. , 0.43826464],
[0. , 0.74198114, 0. , 0. ],
[0. , 3.36221251, 0.23700581, 0.16210278],
[0. , 0. , 0.83725855, 0. ],
[0.13628598, 0.35140916, 1.05812391, 0.27382559],
[1.8598765 , 1.25615814, 0.8924981 , 0.6721845 ],
[0. , 0.17505724, 0. , 2.35889359]])
data[names!='Joe'] =7
data
array([[7. , 7. , 7. , 7. ],
[0. , 0.74198114, 0. , 0. ],
[7. , 7. , 7. , 7. ],
[7. , 7. , 7. , 7. ],
[7. , 7. , 7. , 7. ],
[1.8598765 , 1.25615814, 0.8924981 , 0.6721845 ],
[0. , 0.17505724, 0. , 2.35889359]])
#花式索引：利用整数数组进行索引
arr = np.empty((8,4))
for i in range(8):
arr[i] = i
arr
array([[0., 0., 0., 0.],
[1., 1., 1., 1.],
[2., 2., 2., 2.],
[3., 3., 3., 3.],
[4., 4., 4., 4.],
[5., 5., 5., 5.],
[6., 6., 6., 6.],
[7., 7., 7., 7.]])
arr[[4,3,0,6]]
array([[4., 4., 4., 4.],
[3., 3., 3., 3.],
[0., 0., 0., 0.],
[6., 6., 6., 6.]])
arr[[-3,-5,-7]]
array([[5., 5., 5., 5.],
[3., 3., 3., 3.],
[1., 1., 1., 1.]])
arr = np.arange(32).reshape((8,4))
arr
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]])
arr[[1,5,7,2],[0,3,1,2]]
array([ 4, 23, 29, 10])
arr[[1,5,7,2]][:,[0,3,1,2]]
array([[ 4, 7, 5, 6],
[20, 23, 21, 22],
[28, 31, 29, 30],
[ 8, 11, 9, 10]])
#数组转置和轴对换
arr = np.arange(15).reshape((3,5))
arr
array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14]])
arr.T #转置
array([[ 0, 5, 10],
[ 1, 6, 11],
[ 2, 7, 12],
[ 3, 8, 13],
[ 4, 9, 14]])
arr = np.random.randn(6,3)
arr
array([[ 0.57590651, -0.96882102, 0.30488308],
[-0.77995449, 0.22679996, -0.13410207],
[-0.16624172, -0.38627374, -0.18333949],
[-0.10665208, -0.20768114, 0.86710186],
[ 0.68759888, 0.60503602, 1.19940246],
[-1.35185485, -1.43493499, 0.77248104]])
np.dot(arr.T,arr)# 矩阵内积
array([[ 3.27931205e+00, 1.70736636e+00, -1.39633832e-03],
[ 1.70736636e+00, 3.60749825e+00, -8.17831318e-01],
[-1.39633832e-03, -8.17831318e-01, 2.93170926e+00]])
arr = np.arange(16).reshape((2,2,4))
arr
array([[[ 0, 1, 2, 3],
[ 4, 5, 6, 7]],
[[ 8, 9, 10, 11],
[12, 13, 14, 15]]])
arr.transpose((1,0,2)) #由轴编号组成的元组进行转置
array([[[ 0, 1, 2, 3],
[ 8, 9, 10, 11]],
[[ 4, 5, 6, 7],
[12, 13, 14, 15]]])
arr
array([[[ 0, 1, 2, 3],
[ 4, 5, 6, 7]],
[[ 8, 9, 10, 11],
[12, 13, 14, 15]]])
arr[1,0,0]
8
arr.transpose((1,0,2))[0,1,0]
8
#通用函数：快速的元素及数组函数
arr = np.arange(10)
arr
array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
np.sqrt(arr) #开方
array([0. , 1. , 1.41421356, 1.73205081, 2. ,
2.23606798, 2.44948974, 2.64575131, 2.82842712, 3. ])
np.exp(arr)#e的次幂
array([1.00000000e+00, 2.71828183e+00, 7.38905610e+00, 2.00855369e+01,
5.45981500e+01, 1.48413159e+02, 4.03428793e+02, 1.09663316e+03,
2.98095799e+03, 8.10308393e+03])
x = np.random.randn(8)
y = np.random.randn(8)
x
array([-0.28976486, 0.42334561, -0.57539806, 0.81133112, 0.07536876,
-0.55492229, -0.63633276, 1.02893448])
y
array([-1.52818761, 1.94276598, 1.98097405, -1.0865869 , -1.67389442,
-0.34791017, -1.14349306, -1.15925318])
np.maximum(x,y)
array([-0.28976486, 1.94276598, 1.98097405, 0.81133112, 0.07536876,
-0.34791017, -0.63633276, 1.02893448])
arr = np.random.randn(7)*5
arr
array([ 2.81359527, 8.9677541 , -5.19797091, 2.87857033, -3.685712 ,
7.95114911, 8.16007521])
remainder,whole_part = np.modf(arr)
remainder
array([ 0.81359527, 0.9677541 , -0.19797091, 0.87857033, -0.685712 ,
0.95114911, 0.16007521])
whole_part
array([ 2., 8., -5., 2., -3., 7., 8.])
arr
array([ 2.81359527, 8.9677541 , -5.19797091, 2.87857033, -3.685712 ,
7.95114911, 8.16007521])
np.sqrt(arr)
C:\Anaconda\lib\site-packages\ipykernel_launcher.py:1: RuntimeWarning: invalid value encountered in sqrt
"""Entry point for launching an IPython kernel.
array([1.6773775 , 2.99462086, nan, 1.696635 , nan,
2.8197782 , 2.85658454])
#利用数组进行数据处理
#用数组表达式代替循环的做法，通常被称为矢量化
points = np.arange(-5,5,0.01)
xs,ys = np.meshgrid(points,points)
xs
array([[-5. , -4.99, -4.98, ..., 4.97, 4.98, 4.99],
[-5. , -4.99, -4.98, ..., 4.97, 4.98, 4.99],
[-5. , -4.99, -4.98, ..., 4.97, 4.98, 4.99],
...,
[-5. , -4.99, -4.98, ..., 4.97, 4.98, 4.99],
[-5. , -4.99, -4.98, ..., 4.97, 4.98, 4.99],
[-5. , -4.99, -4.98, ..., 4.97, 4.98, 4.99]])
ys
array([[-5. , -5. , -5. , ..., -5. , -5. , -5. ],
[-4.99, -4.99, -4.99, ..., -4.99, -4.99, -4.99],
[-4.98, -4.98, -4.98, ..., -4.98, -4.98, -4.98],
...,
[ 4.97, 4.97, 4.97, ..., 4.97, 4.97, 4.97],
[ 4.98, 4.98, 4.98, ..., 4.98, 4.98, 4.98],
[ 4.99, 4.99, 4.99, ..., 4.99, 4.99, 4.99]])
z = np.sqrt(xs**2+ys**2)
z
array([[7.07106781, 7.06400028, 7.05693985, ..., 7.04988652, 7.05693985,
7.06400028],
[7.06400028, 7.05692568, 7.04985815, ..., 7.04279774, 7.04985815,
7.05692568],
[7.05693985, 7.04985815, 7.04278354, ..., 7.03571603, 7.04278354,
7.04985815],
...,
[7.04988652, 7.04279774, 7.03571603, ..., 7.0286414 , 7.03571603,
7.04279774],
[7.05693985, 7.04985815, 7.04278354, ..., 7.03571603, 7.04278354,
7.04985815],
[7.06400028, 7.05692568, 7.04985815, ..., 7.04279774, 7.04985815,
7.05692568]])
import matplotlib.pyplot as plt
plt.imshow(z,cmap = plt.cm.gray);plt.colorbar()
plt.title("Image plot of $\sqrt{X^2+y^2}$for a grid of values")
Text(0.5,1,'Image plot of $\\sqrt{X^2+y^2}$for a grid of values')
#将条件逻辑表述为数组运算
xarr = np.array([1.1,1.2,1.3,1.4,1.5])
yarr = np.array([2.1, 2.2, 2.3, 2.4, 2.5])
cond = np.array([True,False,True,True,False])
result = [(x if c else y)
for x,y,c in zip(xarr,yarr,cond)]
result
[1.1, 2.2, 1.3, 1.4, 2.5]
result = np.where(cond,xarr,yarr) #where通常用于根据另一个数组而产生一个新的数组
result
array([1.1, 2.2, 1.3, 1.4, 2.5])
arr = np.random.randn(4,4)
arr
array([[ 0.10412611, -0.51292651, 0.77813603, 0.44210525],
[-0.98470491, 0.50235021, -0.00913422, 1.13403682],
[-1.26298176, 0.48842982, 1.06330363, -1.13924306],
[-0.00624423, 1.52905497, 1.05579894, 0.84535021]])
arr > 0
array([[ True, False, True, True],
[False, True, False, True],
[False, True, True, False],
[False, True, True, True]])
np.where(arr>0,2,-2)
array([[ 2, -2, 2, 2],
[-2, 2, -2, 2],
[-2, 2, 2, -2],
[-2, 2, 2, 2]])
#数学和统计方法
arr = np.random.randn(5,4)
arr
array([[ 0.22053085, -0.40738165, 0.88842433, 0.3305352 ],
[-0.23242076, 0.45397016, -1.62155422, -0.22522925],
[-1.25502714, -1.7575046 , 0.79432674, 0.31165767],
[-1.83870224, 2.47258574, -0.22381713, -0.48870325],
[ 0.83786984, -0.11540301, -1.2092019 , 1.36605066]])
arr.mean()
-0.0849496980348077
np.mean(arr)
-0.0849496980348077
arr.sum()
-1.698993960696154
arr.mean(axis = 1)#行的平均值
array([ 0.25802718, -0.40630852, -0.47663683, -0.01965922, 0.2198289 ])
arr.mean(axis = 0)#列的平均值
array([-0.45354989, 0.12925333, -0.27436444, 0.25886221])
arr = np.array([0,1,2,3,4,5,6,7])#累加函数
arr.cumsum()
array([ 0, 1, 3, 6, 10, 15, 21, 28], dtype=int32)
arr = np.array([[0,1,2],[3,4,5],[6,7,8]])
arr
array([[0, 1, 2],
[3, 4, 5],
[6, 7, 8]])
arr.cumsum(axis =0)
array([[ 0, 1, 2],
[ 3, 5, 7],
[ 9, 12, 15]], dtype=int32)
arr.cumsum(axis = 1)
array([[ 0, 1, 3],
[ 3, 7, 12],
[ 6, 13, 21]], dtype=int32)
arr.cumprod(axis =1)
array([[ 0, 0, 0],
[ 3, 12, 60],
[ 6, 42, 336]], dtype=int32)
#用于布尔型数组的方法
arr = np.random.randn(100)
(arr>0).sum()
58
bools = np.array([False,False,True,False])
bools.any()
True
bools.all()
False
#排序
arr = np.random.randn(6)
arr
array([-1.96112316, -1.44323441, 1.3625597 , -0.52320759, 1.53728505,
-0.39588808])
arr.sort()
arr
array([-1.96112316, -1.44323441, -0.52320759, -0.39588808, 1.3625597 ,
1.53728505])
arr = np.random.randn(5,3)
arr
array([[ 0.77254308, 0.06027608, 1.39994773],
[-0.76576284, -0.3919819 , -0.01057693],
[ 0.07445436, -1.22281584, -2.22202905],
[ 0.55858013, -0.31947992, 1.86954077],
[-0.57716987, -0.04480236, 0.81887336]])
arr.sort(0)
arr
array([[-0.76576284, -1.22281584, -2.22202905],
[-0.57716987, -0.3919819 , -0.01057693],
[ 0.07445436, -0.31947992, 0.81887336],
[ 0.55858013, -0.04480236, 1.39994773],
[ 0.77254308, 0.06027608, 1.86954077]])
arr.sort(1)
arr
array([[-2.22202905, -1.22281584, -0.76576284],
[-0.57716987, -0.3919819 , -0.01057693],
[-0.31947992, 0.07445436, 0.81887336],
[-0.04480236, 0.55858013, 1.39994773],
[ 0.06027608, 0.77254308, 1.86954077]])
#计算分位点
large_arr = np.random.randn(1000)
large_arr.sort()
large_arr[int(0.05*len(large_arr))]
-1.610576889936997
#唯一化以及其他的集合逻辑
names = np.array(['Bob', 'Joe', 'Will', 'Bob', 'Will', 'Joe', 'Joe'])
np.unique(names)
array(['Bob', 'Joe', 'Will'], dtype='<U4')
ints = np.array([3, 3, 3, 2, 2, 1, 1, 4, 4])
np.unique(ints)
array([1, 2, 3, 4])
values = np.array([6,0,0,3,2,5,6])
np.in1d(values,[2,3,6])
array([ True, False, False, True, True, False, True])
#线性代数
x = np.array([[1,2,3],[4,5,6]])
y = np.array([[6,23],[-1,7],[8,9]])
x
array([[1, 2, 3],
[4, 5, 6]])
y
array([[ 6, 23],
[-1, 7],
[ 8, 9]])
x.dot(y) #矩阵相乘
array([[ 28, 64],
[ 67, 181]])
np.dot(x,y)
array([[ 28, 64],
[ 67, 181]])
np.dot(x,np.ones(3))
array([ 6., 15.])
x @ np.ones(3)
array([ 6., 15.])
from numpy.linalg import inv,qr
X = np.random.randn(5,5)
mat = X.T.dot(X)
inv(mat) #方阵的逆
array([[ 0.23995203, 0.09584201, -0.11028433, -0.06880146, 0.21706545],
[ 0.09584201, 0.35099217, 0.68796811, -0.18830081, 0.42573955],
[-0.11028433, 0.68796811, 4.43017504, -0.85985777, 1.26246405],
[-0.06880146, -0.18830081, -0.85985777, 0.35949171, -0.47645408],
[ 0.21706545, 0.42573955, 1.26246405, -0.47645408, 1.02985996]])
mat.dot(inv(mat))
array([[ 1.00000000e+00, -3.73353007e-17, 1.19681218e-17,
2.44722914e-17, -1.72731253e-16],
[-3.36972642e-17, 1.00000000e+00, 2.26347150e-16,
3.31393652e-16, 4.94262742e-16],
[-1.21880052e-16, -6.97956368e-17, 1.00000000e+00,
1.75476227e-16, -6.57906134e-17],
[ 1.31407267e-16, 2.99294724e-16, 9.77143993e-16,
1.00000000e+00, 3.73710817e-16],
[ 1.84787090e-16, 2.94875490e-17, 2.18548932e-16,
-1.57074163e-16, 1.00000000e+00]])
773
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
