Combine numpy arrays to list if first index value is similar

How to combine arrays inside 2D array to a list if the first index value is the same?

Say for example, this 2D array:

[[0 0]
 [0 3]
 [0 4]
 [1 1]
 [2 2]
 [3 0]
 [3 3]
 [3 4]
 [4 0]
 [4 3]
 [4 4]]

How will I make it to something like this?

[[0, 0, 3, 4], [1 1], [2 2], [3 0 3 4], [4 0 3 4]]

When converting from numpy to list, I need it to be optimized as there are thousands of rows on my end.

Any suggestion is much appreciated. The end goal is I want the second index value to be combined together.

Also, take into consideration if ever the first index value is not in ascending order.

>Solution :

You can do it like that:

my_list = [
    [0, 0],
    [0, 3],
    [0, 4],
    [1, 1],
    [2, 2],
    [3, 0],
    [3, 3],
    [3, 4],
    [4, 0],
    [4, 3],
    [4, 4]
    ]

from collections import defaultdict

d = defaultdict(list)
for i, j in my_list:
    d[i].append(j)

combined = [[i]+l for i,l in d.items()]