2. numpy的使用

技术2022-07-10 94

import numpy as np

numpy的基本操作

a = np.array([[1,2,3],[4,5,6]]) b = np.array([1,2,3,4]) c = np.array([[[1,2,3],[4,5,6]],[[1,2,3],[4,5,6]]]) print(a.shape) print(b.shape) print(c.shape)

(2, 3) (4,) (2, 2, 3)

a.dtype

dtype('int64')

a.itemsize

a.size

c.ndim

生成数组

d = np.array([[1,2,3],[4,5,6]],dtype=np.float32) d.dtype

dtype('float32')

e=np.zeros([3,4],dtype="float32") e

array([[0., 0., 0., 0.], [0., 0., 0., 0.], [0., 0., 0., 0.]], dtype=float32)

e.dtype

dtype('float32')

e=np.zeros([3,4],dtype=np.float32) e

array([[0., 0., 0., 0.], [0., 0., 0., 0.], [0., 0., 0., 0.]], dtype=float32)

f=np.ones([3,4],dtype=np.int32) f

array([[1, 1, 1, 1], [1, 1, 1, 1], [1, 1, 1, 1]], dtype=int32)

# 从现有数组生成 score = np.array([[80, 89, 86, 67, 79], [78, 97, 89, 67, 81], [90, 94, 78, 67, 74], [91, 91, 90, 67, 69], [76, 87, 75, 67, 86], [70, 79, 84, 67, 84], [94, 92, 93, 67, 64], [86, 85, 83, 67, 80]]) score

array([[80, 89, 86, 67, 79], [78, 97, 89, 67, 81], [90, 94, 78, 67, 74], [91, 91, 90, 67, 69], [76, 87, 75, 67, 86], [70, 79, 84, 67, 84], [94, 92, 93, 67, 64], [86, 85, 83, 67, 80]])

score1=np.array(score) score2=np.asarray(score) score3=np.copy(score)

score[0][0]=100

score

array([[100, 89, 86, 67, 79], [ 78, 97, 89, 67, 81], [ 90, 94, 78, 67, 74], [ 91, 91, 90, 67, 69], [ 76, 87, 75, 67, 86], [ 70, 79, 84, 67, 84], [ 94, 92, 93, 67, 64], [ 86, 85, 83, 67, 80]])

score2

array([[100, 89, 86, 67, 79], [ 78, 97, 89, 67, 81], [ 90, 94, 78, 67, 74], [ 91, 91, 90, 67, 69], [ 76, 87, 75, 67, 86], [ 70, 79, 84, 67, 84], [ 94, 92, 93, 67, 64], [ 86, 85, 83, 67, 80]])

# 生成固定范围的数组 np.linspace(0, 100, 10)

array([ 0. , 11.11111111, 22.22222222, 33.33333333, 44.44444444, 55.55555556, 66.66666667, 77.77777778, 88.88888889, 100. ])

np.arange(10, 50, 15)

array([10, 25, 40])

a=np.random.uniform(0,100,1000000)

import matplotlib.pyplot as plt plt.figure(figsize=(20,8),dpi=80) plt.hist(a) plt.show()

b=np.random.normal(0,1,(100,100)) b

array([[-6.54070831e-01, -8.58237690e-01, 1.12840828e-03, ..., -5.69480307e-02, 3.55998992e-01, -4.75176936e-01], [ 6.76076814e-01, 1.47650320e+00, 4.61246907e-01, ..., 2.76736091e+00, 2.13077009e-01, -3.92740201e-01], [ 6.14808294e-01, -1.56088314e-01, 8.86520985e-03, ..., 1.47150511e+00, 4.02858641e-01, 2.24224565e+00], ..., [-1.26753533e+00, -1.46173745e-01, -1.42753958e+00, ..., -8.50436456e-02, 5.38332925e-01, -1.05123949e+00], [-7.50317584e-02, -4.60182497e-01, 2.16468102e-01, ..., -1.38952087e-01, -4.67473834e-01, 1.08061128e+00], [-2.42025812e-01, -9.83338435e-02, 1.08822672e-01, ..., -1.26860857e+00, 3.39079875e-01, 2.23547985e-01]])

plt.figure(figsize=(20,8),dpi=80) plt.hist(b,100) plt.show()

数组的基本操作

# 创建符合正态分布的8只股票10天的涨跌幅数据 stock_change = np.random.normal(0, 1, (8, 10)) stock_change

array([[-0.5889853 , -0.93535336, -0.0675299 , -0.72403031, -0.86873673, 0.2366924 , 0.91934989, 0.95733701, 0.7036584 , -0.25490962], [-0.99235101, 0.70940651, -0.66934701, -0.33760595, 1.75578837, -0.17527665, 1.127985 , 1.16480656, 0.60324222, 2.220063 ], [-1.3326412 , -1.01831572, -1.09038797, 0.35145548, -1.52407572, 1.24380911, 0.53576183, -0.87476354, 0.94061172, -1.54695238], [-0.6141568 , 0.34004292, 0.53421929, 0.93273841, -2.39181704, 0.73553393, -0.37974303, 0.12472411, 0.31243104, -0.09388522], [ 0.63699391, 0.54513968, -1.45114066, 0.12302606, 0.8179262 , -0.98861413, -1.37164055, 0.04710424, 0.35452202, -1.77008598], [ 0.05821616, -0.10157081, -0.0177392 , -0.80556969, -0.644111 , 1.62028286, 0.12565102, 0.83290996, 1.06604483, 0.88367563], [ 1.82724499, -0.625139 , -1.00688955, -1.31627874, -0.81109591, 0.96967537, 0.75956849, -0.80162364, 0.44827989, 2.05341272], [-1.7788185 , 0.79750866, -0.29012385, 0.60888444, -1.05643106, 0.17126915, -2.24288296, 0.30024122, -0.44027555, 2.33346992]])

stock_change[0, 0:3]

array([-0.5889853 , -0.93535336, -0.0675299 ])

stock_change.reshape((10,8))

array([[-0.5889853 , -0.93535336, -0.0675299 , -0.72403031, -0.86873673, 0.2366924 , 0.91934989, 0.95733701], [ 0.7036584 , -0.25490962, -0.99235101, 0.70940651, -0.66934701, -0.33760595, 1.75578837, -0.17527665], [ 1.127985 , 1.16480656, 0.60324222, 2.220063 , -1.3326412 , -1.01831572, -1.09038797, 0.35145548], [-1.52407572, 1.24380911, 0.53576183, -0.87476354, 0.94061172, -1.54695238, -0.6141568 , 0.34004292], [ 0.53421929, 0.93273841, -2.39181704, 0.73553393, -0.37974303, 0.12472411, 0.31243104, -0.09388522], [ 0.63699391, 0.54513968, -1.45114066, 0.12302606, 0.8179262 , -0.98861413, -1.37164055, 0.04710424], [ 0.35452202, -1.77008598, 0.05821616, -0.10157081, -0.0177392 , -0.80556969, -0.644111 , 1.62028286], [ 0.12565102, 0.83290996, 1.06604483, 0.88367563, 1.82724499, -0.625139 , -1.00688955, -1.31627874], [-0.81109591, 0.96967537, 0.75956849, -0.80162364, 0.44827989, 2.05341272, -1.7788185 , 0.79750866], [-0.29012385, 0.60888444, -1.05643106, 0.17126915, -2.24288296, 0.30024122, -0.44027555, 2.33346992]])

stock_change.T

array([[-0.5889853 , -0.99235101, -1.3326412 , -0.6141568 , 0.63699391, 0.05821616, 1.82724499, -1.7788185 ], [-0.93535336, 0.70940651, -1.01831572, 0.34004292, 0.54513968, -0.10157081, -0.625139 , 0.79750866], [-0.0675299 , -0.66934701, -1.09038797, 0.53421929, -1.45114066, -0.0177392 , -1.00688955, -0.29012385], [-0.72403031, -0.33760595, 0.35145548, 0.93273841, 0.12302606, -0.80556969, -1.31627874, 0.60888444], [-0.86873673, 1.75578837, -1.52407572, -2.39181704, 0.8179262 , -0.644111 , -0.81109591, -1.05643106], [ 0.2366924 , -0.17527665, 1.24380911, 0.73553393, -0.98861413, 1.62028286, 0.96967537, 0.17126915], [ 0.91934989, 1.127985 , 0.53576183, -0.37974303, -1.37164055, 0.12565102, 0.75956849, -2.24288296], [ 0.95733701, 1.16480656, -0.87476354, 0.12472411, 0.04710424, 0.83290996, -0.80162364, 0.30024122], [ 0.7036584 , 0.60324222, 0.94061172, 0.31243104, 0.35452202, 1.06604483, 0.44827989, -0.44027555], [-0.25490962, 2.220063 , -1.54695238, -0.09388522, -1.77008598, 0.88367563, 2.05341272, 2.33346992]])

stock_change.dtype

dtype('float64')

stock_change.astype(np.float32)

array([[-0.5889853 , -0.93535334, -0.0675299 , -0.7240303 , -0.86873674, 0.2366924 , 0.9193499 , 0.957337 , 0.7036584 , -0.25490963], [-0.992351 , 0.7094065 , -0.669347 , -0.33760595, 1.7557883 , -0.17527665, 1.127985 , 1.1648066 , 0.6032422 , 2.220063 ], [-1.3326412 , -1.0183157 , -1.090388 , 0.35145548, -1.5240757 , 1.2438091 , 0.53576183, -0.87476355, 0.9406117 , -1.5469524 ], [-0.6141568 , 0.34004292, 0.53421926, 0.9327384 , -2.391817 , 0.73553395, -0.37974304, 0.12472411, 0.31243104, -0.09388521], [ 0.6369939 , 0.5451397 , -1.4511406 , 0.12302607, 0.8179262 , -0.98861414, -1.3716406 , 0.04710424, 0.35452202, -1.7700859 ], [ 0.05821616, -0.10157081, -0.0177392 , -0.8055697 , -0.644111 , 1.6202829 , 0.12565102, 0.83290994, 1.0660448 , 0.88367563], [ 1.827245 , -0.625139 , -1.0068896 , -1.3162787 , -0.8110959 , 0.96967536, 0.7595685 , -0.80162364, 0.4482799 , 2.0534127 ], [-1.7788185 , 0.79750866, -0.29012385, 0.60888445, -1.056431 , 0.17126915, -2.242883 , 0.3002412 , -0.44027555, 2.3334699 ]], dtype=float32)

数组的运算

# 逻辑运算 stock_change stock_change1=np.copy(stock_change)

stock_change>0.5 # 逻辑判断, 如果涨跌幅大于0.5就标记为True 否则为False：

array([[False, False, False, False, False, False, True, True, True, False], [False, True, False, False, True, False, True, True, True, True], [False, False, False, False, False, True, True, False, True, False], [False, False, True, True, False, True, False, False, False, False], [ True, True, False, False, True, False, False, False, False, False], [False, False, False, False, False, True, False, True, True, True], [ True, False, False, False, False, True, True, False, False, True], [False, True, False, True, False, False, False, False, False, True]])

stock_change1[stock_change1>0.5]=1 # 如果涨跌幅大于0.5就把值改为1

stock_change1

array([[-0.5889853 , -0.93535336, -0.0675299 , -0.72403031, -0.86873673, 0.2366924 , 1. , 1. , 1. , -0.25490962], [-0.99235101, 1. , -0.66934701, -0.33760595, 1. , -0.17527665, 1. , 1. , 1. , 1. ], [-1.3326412 , -1.01831572, -1.09038797, 0.35145548, -1.52407572, 1. , 1. , -0.87476354, 1. , -1.54695238], [-0.6141568 , 0.34004292, 1. , 1. , -2.39181704, 1. , -0.37974303, 0.12472411, 0.31243104, -0.09388522], [ 1. , 1. , -1.45114066, 0.12302606, 1. , -0.98861413, -1.37164055, 0.04710424, 0.35452202, -1.77008598], [ 0.05821616, -0.10157081, -0.0177392 , -0.80556969, -0.644111 , 1. , 0.12565102, 1. , 1. , 1. ], [ 1. , -0.625139 , -1.00688955, -1.31627874, -0.81109591, 1. , 1. , -0.80162364, 0.44827989, 1. ], [-1.7788185 , 1. , -0.29012385, 1. , -1.05643106, 0.17126915, -2.24288296, 0.30024122, -0.44027555, 1. ]])

stock_change

np.all(stock_change[0:2,0:5]>0)

False

np.any(stock_change[0:2,0:5]>0)

True

temp=stock_change[0:4,0:4] np.where(temp>0,1,0)

array([[0, 0, 0, 0], [0, 1, 0, 0], [0, 0, 0, 1], [0, 1, 1, 1]])

np.where(np.logical_and(temp > 0.5, temp < 1), 1, 0)

array([[0, 0, 0, 0], [0, 1, 0, 0], [0, 0, 0, 0], [0, 0, 1, 1]])

np.where(np.logical_or(temp > 0.5, temp < -0.5), 1, 0)

array([[1, 1, 0, 1], [1, 1, 1, 0], [1, 1, 1, 0], [1, 0, 1, 1]])

temp

array([[-0.5889853 , -0.93535336, -0.0675299 , -0.72403031], [-0.99235101, 0.70940651, -0.66934701, -0.33760595], [-1.3326412 , -1.01831572, -1.09038797, 0.35145548], [-0.6141568 , 0.34004292, 0.53421929, 0.93273841]])

np.max(temp)

0.9327384109856901

np.max(temp,axis=1)

array([-0.0675299 , 0.70940651, 0.35145548, 0.93273841])

np.max(temp, axis=0)

array([-0.5889853 , 0.70940651, 0.53421929, 0.93273841])

np.std(temp,axis=1)

array([0.320048 , 0.63915226, 0.65926327, 0.56843862])

np.argmax(temp,axis=1)

array([2, 1, 3, 3])

数组间的运算

arr=np.array([[1,2,3,4],[5,6,7,8]])

arr

array([[1, 2, 3, 4], [5, 6, 7, 8]])

arr+10

array([[11, 12, 13, 14], [15, 16, 17, 18]])

arr*10

array([[10, 20, 30, 40], [50, 60, 70, 80]])

arr/10

array([[0.1, 0.2, 0.3, 0.4], [0.5, 0.6, 0.7, 0.8]])

### 矩阵的计算 a = np.array([[80, 86], [82, 80], [85, 78], [90, 90], [86, 82], [82, 90], [78, 80], [92, 94]]) b = np.array([[0.7], [0.3]])

a_mat=np.mat(a)

a_mat

matrix([[80, 86], [82, 80], [85, 78], [90, 90], [86, 82], [82, 90], [78, 80], [92, 94]])

np.matmul(a,b)

array([[81.8], [81.4], [82.9], [90. ], [84.8], [84.4], [78.6], [92.6]])

np.dot(a,b)

array([[81.8], [81.4], [82.9], [90. ], [84.8], [84.4], [78.6], [92.6]])

a@b

array([[81.8], [81.4], [82.9], [90. ], [84.8], [84.4], [78.6], [92.6]])

b_mat=np.mat(b)

a_mat*b_mat

matrix([[81.8], [81.4], [82.9], [90. ], [84.8], [84.4], [78.6], [92.6]])

Processed: 0.016, SQL: 9