16/12/22, 12:59 Sesión 05.

ipynb - Colaboratory

Matrices

import numpy as np

A=np.array([1,2,3,5-1,])

type(A)

numpy.ndarray

A[0]

A[[1,2]]

array([2, 3])

Manejo de indices

A.dtype

dtype('int64')

L=list(A)

L[0]

L[[1,3]]

---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-9-2c640c2d1580> in <module>
----> 1 L[[1,3]]

TypeError: list indices must be integers or slices, not list

SEARCH STACK OVERFLOW

L[1:3]

[2, 3]

A[A>0]

array([1, 2, 3, 4])

A[-1]

A[-2]

A>0

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

np.logical_and(A>0,A<3)

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

A[np.logical_and(A>0,A<3)]

array([1, 2])

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 1/8
16/12/22, 12:59 Sesión 05.ipynb - Colaboratory

(A>0) & (A<3)

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

(A>0).all()

True

(A>0).any()

True

Redimensionamiento

A=np.array([1,2,3,-2])

A=A.reshape((2,2))

print(A)

[[ 1 2]
[ 3 -2]]

A=A.reshape((2,2),order='F')
print(A)

[[ 1 2]
[ 3 -2]]

A.ndim

A.shape

(2, 2)

A.size

B=np.array([2,3,4,5])
B.shape

(4,)

B=np.array([[2],[3],[4],[5]])
B.shape

(4, 1)

B=np.array([[2,3,4,5]])
B.shape

(1, 4)

Operaciones con matrices

B=np.array([[1,3],[2,1]])

print(B)

[[1 3]
[2 1]]

A+B

array([[ 2, 5],
[ 5, -1]])

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 2/8
16/12/22, 12:59 Sesión 05.ipynb - Colaboratory

A*B

array([[ 1, 6],
[ 6, -2]])

A/B

array([[ 1. , 0.66666667],
[ 1.5 , -2. ]])

B**2

array([[1, 9],
[4, 1]])

np.exp(A)

array([[ 2.71828183, 7.3890561 ],

[20.08553692, 0.13533528]])

np.sin(A)

array([[ 0.84147098, 0.90929743],

[ 0.14112001, -0.90929743]])

np.sqrt(np.abs(A))

array([[1. , 1.41421356],
[1.73205081, 1.41421356]])

C=np.array([1,3])

C[:,None]

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

Inner y dot

a=np.array([1,2])
b=np.array([3,4])

print(np.dot(a,b))

11

print(np.inner(a,b))

11

a=np.array([[1,2],[3,4]])
b=np.array([[11,12],[13,14]])

print(np.dot(a,b))

[[37 40]
[85 92]]

print(np.inner(a,b))

[[35 41]
[81 95]]

print(np.inner(a,b.T))

[[37 40]
[85 92]]

print(a@b)

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 3/8
16/12/22, 12:59 Sesión 05.ipynb - Colaboratory

[[37 40]
[85 92]]

Traza, norma y determinante

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

traza=np.trace(A)
print(traza)

-1

norma=np.linalg.norm(A)
print(norma)

4.242640687119285

det=np.linalg.det(A)
print(det)

-8.000000000000002

Normas matriciales
norma1: max(sum(abs(x),axis=0))

norma2: mayor valor propio en valor absoluto

norma inf: max(sum(abs(x),axis=1))

print(np.linalg.norm(A,1))

4.0

print(np.linalg.norm(A,2))

3.6225827286091983

print(np.linalg.norm(A,np.inf))

5.0

Ejercicio
Verifique el teorema de Cayley Hamilton para una matriz, es decir, si p(x) , entonces p(A)
= det(A − xI ) = 0 .

trace_A=np.trace(A)
det_A=np.linalg.det(A)
I=np.eye(2)
A@A-trace_A*A+det_A*I

array([[-1.77635684e-15, 0.00000000e+00],
[ 0.00000000e+00, -1.77635684e-15]])

Valores propios

from numpy.linalg import eig

eigen_val, eigen_vec=eig(A)

print(eigen_val)

[ 2.37228132 -3.37228132]

print(eigen_vec)

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 4/8
16/12/22, 12:59 Sesión 05.ipynb - Colaboratory

[[ 0.82456484 -0.41597356]
[ 0.56576746 0.90937671]]

Comprobamos que Av = λv

A@eigen_vec[:,0]

array([1.95609977, 1.34215959])

eigen_val[0]*eigen_vec[:,0]

array([1.95609977, 1.34215959])

A@eigen_vec[:,0]-eigen_val[0]*eigen_vec[:,0]

array([2.22044605e-16, 0.00000000e+00])

np.diag(eigen_val)@eigen_vec

array([[ 1.95609977, -0.9868063 ],

[-1.90792706, -3.06667409]])

Ejemplo
Diagonalizar la matriz

A=
1 2 4
⎡ ⎤

⎢2 1 −4 ⎥
⎣ ⎦
0 0 3

es decir encuentre D y P tal que A = P DP

−1
.

A=np.array([[1,2,4],[2,1,-4],[0,0,3]])

eigen_val,eigen_vec=eig(A)

print([email protected](eigen_val)@np.linalg.inv(eigen_vec))

[[ 1. 2. 4.]
[ 2. 1. -4.]
[ 0. 0. 3.]]

print(eigen_vec)

[[ 0.70710678 -0.70710678 0.02616462]

[ 0.70710678 0.70710678 -0.8888268 ]
[ 0. 0. 0.45749571]]

print(np.diag(eigen_val))

[[ 3. 0. 0.]
[ 0. -1. 0.]
[ 0. 0. 3.]]

Ejemplo
Resuelva el sistema
x + 2y = 10

3x + 4y = 20

from numpy.linalg import solve

A=np.array([[1,2],[3,4]])
b=np.array([10,20])

x=solve(A,b)
print(x)

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 5/8
16/12/22, 12:59 Sesión 05.ipynb - Colaboratory

[0. 5.]

Ejercicio
Implemente las transformaciones rigidas como funciones que acepten un array (x, y, 1) y devuelva un (x′ , y ′ , 1) .

1. Traslación
1 0 h1 x
⎡ ⎤⎡ ⎤
Th (x, y) = ⎢ 0 1 h2 ⎥ ⎢ y ⎥
⎣ ⎦⎣ ⎦
0 0 1 1

def T(x,h):
h1=h[0]
h2=h[1]
matT=np.array([[1,0,h1],[0,1,h2],[0,0,1]])
z=matT@x
return z

x=np.array([2,1,1])
h1=np.array([1,3,1])
h2=np.array([2,2,1])
print(T(x,h1))

[3 4 1]

def mT(h):
h1=h[0]
h2=h[1]
matT=np.array([[1,0,h1],[0,1,h2],[0,0,1]])
return matT

mT(h1)@x

array([3, 4, 1])

mT(h2)@x

array([4, 3, 1])

mT(h1)@mT(h2)@x

array([5, 6, 1])

mT(h1+h2)@x

array([5, 6, 1])

2. Rotación
cos θ − sin θ 0 x
⎡ ⎤⎡ ⎤
Rθ (x, y) = ⎢ sin θ cos θ 0⎥⎢y ⎥
⎣ ⎦⎣ ⎦
0 0 1 1

def rotacion(x,theta):
ct=np.cos(theta)
st=np.sin(theta)
matR=np.array([[ct,-st,0],[st,ct,0],[0,0,1]])
z=matR@x
return z

x=np.array([2,1,1])
theta=np.pi/2
print(rotacion(x,theta))

[-1. 2. 1.]

def mR(theta):
ct=np.cos(theta)
st=np.sin(theta)
matR=np.array([[ct,-st,0],[st,ct,0],[0,0,1]])
return matR

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 6/8
16/12/22, 12:59 Sesión 05.ipynb - Colaboratory

mR(np.pi/2)@x

array([-1., 2., 1.])

mR(np.pi/3)@x

array([0.1339746 , 2.23205081, 1. ])

mR(np.pi/3+np.pi/2)@x

array([-2.23205081, 0.1339746 , 1. ])

Ejemplo
Pruebe girar un cuadrado de vértices A(1, 1), B(3, 1), C (3, 3) y D(1, 3) alrededor de un eje perpendicular al plano que pase por P (−2, 2).

figura=np.array([[1,3,3,1,1],[1,1,3,3,1],[1,1,1,1,1]])
print(figura)

[[1 3 3 1 1]
[1 1 3 3 1]
[1 1 1 1 1]]

h=np.array([-2,2,1])
theta=np.pi/3

T1=mT(-h)
R1=mR(theta)
figura1=mT(h)@mR(theta)@mT(-h)@figura
T1

array([[ 1, 0, 2],
[ 0, 1, -2],
[ 0, 0, 1]])

import matplotlib.pyplot as plt

fig=plt.figure()
ax=fig.add_subplot(111)
plt.plot(figura[0,:],figura[1,:],'r')
plt.plot(figura1[0,:],figura1[1,:],'b')
plt.plot(h[0],h[1],'o')
ax.set_aspect('equal',adjustable='box')

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 7/8
16/12/22, 12:59 Sesión 05.ipynb - Colaboratory

check 0 s completado a las 12:59

https://colab.research.google.com/drive/1xl7LnNhPch3VVLfIzr7OZFRDvthf1zrU#scrollTo=uJ5h2rNFEAmo&printMode=true 8/8