numpy tolist() uncomfortable output

Question

I'm trying to extract a column of a numpy matrix in a list form. I've used the method tolist(), but it is not useful for my purpose. Let see the code.

import numpy as np
def get_values(feature):
    '''
    This method creates a lst of all values in a feature, without repetitions
    :param feature: the feature of which we want to extract values
    :return: lst of the values
    '''
    values = []
    for i in feature:
        if i not in values:
            values.append(i)
    return values
lst=[1, 2, 4, 4, 6]
a=get_values(lst)
print(a)
b=np.matrix('1 2; 3 4')
col = b[:,0].tolist()
print(col)
if col == [1, 3]:
    print('done!')

OUTPUT

[1, 2, 4, 6]
[[1], [3]]

As you can see, the returned list from method tolist() is ignored in the if statement. Now, if I can't change the if statement (for any reason) how can I do to manage the b as if it were a list like a?

Answer 1

The problem is that numpy.matrix objects always maintain two dimensions. Convert to array, then flatten:

>>> col = b[:,0].getA().flatten().tolist()
>>> col
[1, 3]

Or maybe just work with normal numpy.ndarrays...

>>> a = b.getA()
>>> a[:,0]
array([1, 3])

vs...

>>> b[:,0]
matrix([[1],
        [3]])