Condition python Matrices

Question

My Code:

 m1 = [[1, 2, 3],
      [4, 5, 6]]
 m2 = [[2, 3, 4],
       [5, 6, 7]]

def soma_matrizes(m1, m2):
       matriz_soma = []
       linhas = len(m1) 
       colunas = len(m1[0]) 
  for i in range(linhas):
       matriz_soma.append([])
       for j in range(colunas):
            soma = m1[i][j] + m2[i][j]
            matriz_soma[i].append(soma)
 return matriz_soma

Example 1:

 m1 = [[1, 2, 3], [4, 5, 6]]
 m2 = [[2, 3, 4], [5, 6, 7]]
 soma_matrizes(m1, m2) => [[3, 5, 7], [9, 11, 13]]

Example 2:

m1 = [[1], [2], [3]]
m2 = [[2, 3, 4], [5, 6, 7]]
soma_matrizes(m1, m2) => False

What I have to do to make my code meet the requirements of the exemplo1 and exemplo2?

Answer 1

The simplest way is to use the Numpy, because there you can compare if the format (shape) of the matrices is different, and if it is to return False as you wish. Example code:

def soma_matrizes(m1, m2):
    import numpy as np
    m1 = np.array(m1)
    m2 = np.array(m2)
    if m1.shape != m2.shape:
        return False
    else:
        return (m1 + m2).tolist()

m1 = [[1, 2, 3],[4, 5, 6]]
m2 = [[2, 3, 4],[5, 6, 7]]  
print(soma_matrizes(m1, m2))

m1 = [[1], [2], [3]]
m2 = [[2, 3, 4],[5, 6, 7]]  
print(soma_matrizes(m1, m2))

Upshot:

[[3, 5, 7], [9, 11, 13]]
False

See working in Ideone.

Editing:

To do manually (considering only lists):

def sameShape(m1, m2):
    '''Verificação manual se duas listas têm o mesmo formato'''

    if type(m1) != type(m2):
        return False

    if type(m1) == list:
        if len(m1) != len(m2):
            return False

        for i in range(len(m1)):
            if not sameShape(m1[i], m2[i]):
                return False

    return True

In this case, remove the use of the Numpy library and change the line:

if m1.shape != m2.shape:

for:

if not sameShape(m1, m2):

But note that you can no longer simply do m1 + m2, because this operator concatenates the Python lists and does not add up. You can use your summation method either, or do so (see in Ideone):

def sameShape(m1, m2):
    '''Verificação manual se duas listas têm o mesmo formato'''

    if type(m1) != type(m2):
        return False

    if type(m1) == list:
        if len(m1) != len(m2):
            return False

        for i in range(len(m1)):
            if not sameShape(m1[i], m2[i]):
                return False

    return True

def soma_matrizes(m1, m2):
    if not sameShape(m1, m2):
        return False
    else:
        # Reduz a dimensão da matriz de 2 pra 1 (i.e. transforma em lista simples)
        # para facilitar a soma
        m1 = [i for j in m1 for i in j]
        m2 = [i for j in m2 for i in j]

        # Calcula a soma item a item das duas listas
        s = [sum(t) for t in zip(m1, m2)] # <= usa `sum` em cada tupla
        #s = [i + j for i, j in zip(m1, m2)] # <== alternativa (talvez mais fácil de entender)

        # Faz a lista de soma ter 2 dimensões antes de retornar
        k = int(len(s) / 2)
        return [s[:k], s[k:]]

m1 = [[1, 2, 3],[4, 5, 6]]
m2 = [[2, 3, 4],[5, 6, 7]]
print(soma_matrizes(m1, m2))

m1 = [[1], [2], [3]]
m2 = [[2, 3, 4],[5, 6, 7]]
print(soma_matrizes(m1, m2))

Finally, note that:

1. This code is much bigger and harder to understand than the previous one, with Numpy.
2. It is specific to a 2-dimensional matrix and would need to be adapted if its matrix had more dimensions. Numpy’s code is already in general use.

Therefore, I disagree of his understanding that without the use of the package is better (even for you). : ) It is not so difficult to install it and worth very sorry! Believe me.