Scipy 习题

Exercise 10.1: Least squares

Generate matrix A ∈ Rm×n with m > n. Also generate some vector b ∈ R^m.

Now find x = arg minx ||Ax − b||2.

Print the norm of the residual.

from scipy.optimize import least_squares  # 引入最小二乘函数
import numpy as np


m = 20
n = 5
A = np.random.rand(m, n)
b = np.random.rand(m, 1)
x = np.zeros(n)


def func(x):
    return np.linalg.norm(A.dot(x) - b)


sol = least_squares(func, x)
print(sol)
print("argmin x :", sol.x)
print(" residual：", func(sol.x))

输出：

 active_mask: array([0., 0., 0., 0., 0.])
        cost: 18.548562450660654
         fun: array([6.09074092])
        grad: array([-0.00043129,  0.00825182,  0.0032902 , -0.00024505,  0.0037066 ])
         jac: array([[-7.08103180e-05,  1.35481358e-03,  5.40196896e-04,
        -4.02331352e-05,  6.08563423e-04]])
     message: '`ftol` termination condition is satisfied.'
        nfev: 117
        njev: 113
  optimality: 0.0082518184788379
      status: 2
     success: True
           x: array([ 0.12208824, -0.06974677,  0.27798636,  0.32015007,  0.30762308])
 argmin x : [ 0.12208824 -0.06974677  0.27798636  0.32015007  0.30762308]
 residual： 6.090740915629338

Exercise 10.2: Optimization

Find the maximum of the function

import math
from scipy.optimize import fmin


def func(x):
    return ((math.sin(x - 2))**2) * math.exp(-(x**2))


result = fmin(func, 0)
print(result)

输出：

Optimization terminated successfully.
         Current function value: 0.000000
         Iterations: 25
         Function evaluations: 50
[-1.1415625]

Exercise 10.3: Pairwise distances

Let X be a matrix with n rows and m columns. How can you compute the pairwise distances between every two rows?

As an example application, consider n cities, and we are given their coordinates in two columns. Now we want a nice table that tells us for each two cities, how far they are apart.

Again, make sure you make use of Scipy's functionality instead of writing your own routine.

import numpy as np
from scipy.spatial import distance
# http://www.it1352.com/234917.html

m = 20
n = 5

X = np.random.rand(m, n)
dists = distance.pdist(X)
print(dists)

输出：

[1.02397042 0.66198184 0.68270417 0.95980503 0.94876295 1.28910145
 0.99786154 1.19195332 0.68494725 0.59830461 0.58438626 1.00953439
 1.1914291  1.0226026  1.00228754 0.8516124  1.19131237 0.69077538
 0.719278   0.95692451 0.95817779 0.39873196 0.64805639 0.71334487
 1.19141197 0.69623684 0.76592437 0.74263162 0.81217497 0.57775737
 0.56039256 0.61688712 0.93791503 0.67362314 0.6467085  0.83132238
 0.95854108 0.91257872 1.12184453 1.11758816 1.29202565 1.3037946
 1.28226153 0.82495697 0.43554784 0.92698746 1.013793   1.21732681
 1.13076661 1.09218133 1.10996813 1.35308725 0.82482311 0.77525791
 1.01756844 0.89849962 1.14664437 0.57192909 1.09120386 0.63602684
 0.5599489  0.78490047 0.92842496 1.0890859  1.08613869 1.03776629
 0.59579914 1.08663416 0.17079976 0.84733913 0.59838276 0.75257815
 1.16063814 0.64251211 0.7746636  0.92077766 0.65463017 0.69414432
 0.58625837 0.52743924 0.9424456  0.60484429 0.56911501 0.93706529
 0.99891599 1.01113665 1.07217574 0.84167062 0.78272859 0.84031636
 0.91258276 1.02240877 1.04741499 1.06635777 0.43623997 0.56796122
 0.3607233  0.79260387 0.64134569 1.029979   0.24457069 0.69491635
 1.15515043 0.8260265  1.07687535 0.47372058 0.8340751  1.19018361
 0.71294476 1.07642665 1.0829414  1.19344022 0.95507728 0.64252197
 1.03617533 0.83883661 1.28503487 1.13280558 1.25331462 1.1769534
 0.58440816 1.26014575 0.66493128 1.05428955 0.61345158 1.1251426
 0.74108101 1.10167927 0.5731328  0.85382408 1.01816151 0.57873184
 0.93051386 1.03131713 1.05847336 0.68455967 0.46310971 1.01288352
 0.81169475 0.90925392 0.8417417  0.45930051 1.03363045 0.59979694
 0.65918649 0.82508695 0.77421334 1.05611904 0.99491226 0.9204424
 0.80624712 1.04649568 0.43769581 0.67649178 0.88116259 0.73154967
 0.64295929 1.08572208 0.59676836 1.0783366  0.79180808 0.92484832
 0.69486641 0.49884102 1.35050115 0.90339827 0.97047237 0.85480938
 1.2740189  0.41498608 1.33672543 0.75396756 1.02861397 1.03091838
 1.30615074 1.38120514 0.85513885 1.03696625 1.04190856 1.31548767
 0.82111819 0.7661898  0.93336994 0.39533012 0.71387853 0.56117792
 0.81520453 0.98433292 0.99389997 0.75721045]