Matrices depending on a parameter in python

Question

I've got a matrix which can be defined using a parameter, so this is like for a parameter 'a' I can label my Matrix as A(a) where the entries depend on this 'a', the thing is that I'd like to create a set of this matrices using a domain for this 'a's. So my question is, How do I define the set of 'a' in order to obtain the set of matrices?

Answer 1

Something you can do is using the method fromfunction of the numpy library. Then with a (nxn) square matrix it would be as simple as

import numpy as np

def your_function(row_index, col_index, a):
    # ... do stuff and compute your result
    return result

n = 2
a = np.pi
your_matrix = np.fromfunction(lambda i, j: your_function(i, j, a), (n, n), dtype=np.float)

and if you want it to be in a callable form, actually as you asked, you can do

A = lambda a : np.fromfunction(lambda i, j: your_function(i, j, a), (n, n), dtype=np.float)

Or if you do not want to use an anonymous function

def A(a):
    return  np.fromfunction(lambda i, j: your_function(i, j, a), (n, n), dtype=np.float)

---

Let do a simple example

def your_function(row_index, col_index, a):
    return row_index + col_index + a
n = 2
A = lambda a : np.fromfunction(lambda i, j: your_function(i, j, a), (n, n), dtype=np.float)

Use case

>>> A(1)
array([[ 1.,  2.],
       [ 2.,  3.]])

>>> A(np.pi)
array([[ 3.1415926535897931,  4.1415926535897931],
       [ 4.1415926535897931,  5.1415926535897931]])