NumPy – Matrix Library:

Matrix module contains the functions which return matrices instead of arrays.

matrix – This will return a matrix from an array kind of an object or from the string of data.

Example:

>>> x=np.matrix('1,2,3,4')

>>> x
matrix([[1, 2, 3, 4]])

>>> print(x)
[[1 2 3 4]]

asmatrix – This will interpret the input as matrix.

Example:

>>> x=np.array([[1,2],[3,4],[4,5]])

>>> x
array([[1, 2],
[3, 4],
[4, 5]])

>>> y=np.asmatrix(x)

>>> y
matrix([[1, 2],
[3, 4],
[4, 5]])

>>> x[0,0]=5

>>> y
matrix([[5, 2],
[3, 4],
[4, 5]])

Below are the some replacement functions in matlib.

empty: This will return a new matrix of given shape and type.

Example:

>> import numpy.matlib as mb

# this will generate the random data

>>> mb.empty((2,2))
matrix([[ 4.00535364e-307, 2.33648882e-307],
[ 3.44900029e-307, 1.78250172e-312]])

>>> mb.empty((2,2),int)
matrix([[0, 1],
[2, 3]])

>>> mb.empty((2,2),int)
matrix([[1, 2],
[3, 4]])

>>> mb.empty((2,2),int)
matrix([[0, 1],
[2, 3]])

zeros: This will return a matrix of given shape and type, filled with zeros.

Example:

>>> import numpy.matlib as mb

>>> mb.zeros((3,3))
matrix([[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.]])

>>> mb.zeros((3,3),int)
matrix([[0, 0, 0],
[0, 0, 0],
[0, 0, 0]])

ones: This will return a matrix of ones.

Example:

>>> import numpy.matlib as mb

>>> mb.ones((3,3))
matrix([[ 1., 1., 1.],
[ 1., 1., 1.],
[ 1., 1., 1.]])

>>> mb.ones((3,3),int)
matrix([[1, 1, 1],
[1, 1, 1],
[1, 1, 1]])

eye: This will return a matrix with ones on diagonal and zeros elsewhere.

Example:

>>> import numpy.matlib as mb

>>> mb.eye(3)
matrix([[ 1., 0., 0.],
[ 0., 1., 0.],
[ 0., 0., 1.]])

>>> mb.eye(3,dtype=int)
matrix([[1, 0, 0],
[0, 1, 0],
[0, 0, 1]])

identity: This will return a square identity matrix of given size.

Example:

>>> import numpy.matlib as mb

>>> mb.identity(3,dtype=int)
matrix([[1, 0, 0],
[0, 1, 0],
[0, 0, 1]])

>>> mb.identity(4,dtype=int)
matrix([[1, 0, 0, 0],
[0, 1, 0, 0],
[0, 0, 1, 0],
[0, 0, 0, 1]])

rand: This will return a matrix of random values with given shape.

Example:

>>> import numpy.matlib as mb

>>> mb.rand(2,3)
matrix([[ 0.76296068, 0.20586189, 0.18435659],
[ 0.56268214, 0.47163051, 0.21750408]])

# if the first argument is tuple, Other arguments are ignored.

>>> mb.rand((2,3),int)
matrix([[ 0.34026611, 0.35534349, 0.00456508],
[ 0.32700061, 0.89152015, 0.74071634]])