python numpy作业

最新推荐文章于 2023-04-11 16:35:44 发布

原创最新推荐文章于 2023-04-11 16:35:44 发布 · 336 阅读

0 ·

CC 4.0 BY-SA版权

本文详细介绍了如何使用Python的Numpy库进行数值计算和数据操作。通过实例代码，涵盖了数组创建、数组运算、统计函数应用等方面，帮助读者深入理解Numpy在数据分析中的应用。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

代码：

import numpy
import time
from scipy import linalg

n = 200
m = 500
A = numpy.random.normal(size = (n, m))
f = numpy.random.randint(1, 1000, size = m)
B = linalg.toeplitz(f)

#9.1
#A+A
A_A = A + A
'''print(A_A)'''
#AAT
AAT = numpy.dot(A, A.T)
'''print(AAT)'''
#ATA
ATA = numpy.dot(A.T, A)
'''print(ATA)'''
#AB
AB = numpy.dot(A, B)
'''print(AB)'''
#计算A(B-sI)
def func1(A, B, s):
    I = numpy.eye(B.shape[0])
    return numpy.dot(A, B-s*I)

s = 3
result = func1(A, B, s)
'''print(result)'''

#9.2
b = numpy.random.random(size = m) * 10
#求解线性方程Bx = b
def solve(B, b):
    return numpy.linalg.solve(B, b)
x = solve(B, b)
'''print(x)'''

#9.3
def FA(A):
    return numpy.linalg.norm(A, "fro")
#the Frobenius norm of A:
FofA = FA(A)
'''print(FofA)'''

#the infinity norm of B
def IB(B):
    return numpy.linalg.norm(B, numpy.inf)
IofB = IB(B)
'''print(IofB)'''

#singular values
U, sigma, V = numpy.linalg.svd(B)
maxSV = max(sigma)
minSV = min(sigma)
'''print(maxSV)
print(minSV)'''

#9.4
#求Z的特征值和特征向量

Z = numpy.random.normal(size = (n, n))
#eigenvalue和eigenvecor
def maxabs(x):
    k = 1
    for i in range(len(x)):
        if abs(x[i]) > abs(x[k]):
            k = i
    return abs(x[k])

def max_eigen(x0, Z, eps):
    count = 0

    x1 = x0
    y = x1/maxabs(x1)
    x0 = numpy.dot(Z, x0)
    count += 1
    while abs(maxabs(x1) - maxabs(x0)) > eps and count < 10000:
        x1 = x0
        y = x1/maxabs(x1)
        x0 = numpy.dot(Z, y)
        count += 1

    return maxabs(x1), y, count

x0 = numpy.ones(n)
eps = 0.001
t = time.clock()
e_value1, e_vector1, count = max_eigen(x0, Z, eps)
t = time.clock() - t

'''print(e_value1)
print(e_vector1)
print(count)
print(t)'''

#9.5
C = numpy.zeros((n,n))
p = 0.8
count = 0
for i in range(n):
    for j in range(n):
        if numpy.random.rand() < p:
            C[i,j] = 1
            count += 1


'''print(C)'''
'''print(count)'''

p = count/(n*n)
U1, sigma1, V1 = numpy.linalg.svd(C)
maxSV1 = max(sigma1)

'''print(n)
print(p)
print(maxSV1)
print(n*p)'''
#观察并计算可见maxSV1等于n*p

#9.6
z = 400
arr = numpy.random.randint(1, 1000, size = n)
def nearest(arr, z):
    arr = arr.reshape(arr.size)
    cmp_arr = abs(arr - z)
    return arr[numpy.argmin(cmp_arr)]

nearest_value = nearest(arr, z)
'''print(nearest_value)'''