is there a faster way to create a array of elements of an empty array without using for loops

Question

I want to create an array of elements. I can do it using a loop, but i want to know if there is another way, like an numpy function or something to do this in less timme.

this is what i want to do:

def no_loops(y):


   mat = []
   for i in range(5012):
       mat.append(y[i])

   fun = funcion_a(mat)

   return fun

The y array is empty. The elements of the array are the solutions of an equation system that scipy will solve. I want to change the for loop for something faster.

***y** array is empty* - and how is that `y[i]` working in such case? Does not look like working code — RomanPerekhrest, Nov 11 '19 at 19:35
root from scipy optimize needs an empty array to itarate and find a solution to an non-linear equation system. — Gian Luigi Canepa Sprovera, Nov 11 '19 at 19:38
y dosen't exist, all i want to do its to give the index of the values that already have in another function and solve them. — Gian Luigi Canepa Sprovera, Nov 11 '19 at 19:39

NicholasM · Accepted Answer · 2019-11-11T19:43:14.760

If you want a new array of a certain size, and want to ignore the very confusing y from your sample code, you could do something like this:

def no_loops():
    # Create an array of length 5012, all elements of which are zero.
    return funcion_a(numpy.zeros(5012))

You could also use numpy.empty in this case, but I prefer the determinism of numpy.zeros.

If you want to copy the data in y, you could explicitly copy using numpy.array:

def no_loops(y):
   return funcion_a(numpy.array(y[:5012]))

You could also investigate using numpy.asanyarray if you need to support subclasses of numpy.array. See: Numpy - array vs asarray

is there a faster way to create a array of elements of an empty array without using for loops

1 Answers1