You're welcome!

Glad you like them!

You're welcome!

Thanks!

Because adding two matrices requires traversing all elements and a matrix of size n by n has n² elements

I don't think that these algorithms are asked in an interview, but they're good to know

what kind of algorithms can be asked in an interview, please ?@@insidecode

if n is big enough yes

For sufficiently large matrices (1mil x 1mil - Matrices and beyond), newer algorithms with O(n^2.34) were found These are only worth using on really big matrices though because they have big constant factors in their time complexity that the big O notation does not show. In general, just use the method from numpy

Thanks!

Thanks!

yes because the code works with numpy arrays not normal lists

@@insidecode I used numpy arrays for my input matrices though

Weird, I'll try the code when I get home and I'll see

@@elliebartlett Update: I tried the code and it worked perfectly, I wrote as input m1 = np.array([[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12], [13, 14, 15, 16]]) m2 = np.array([[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12], [13, 14, 15, 16]]) print(strassen(m1, m2))

@@insidecode hi thanks for checking, I tried it again with your exact inputs and I’m still getting the same error. I can’t see a difference in my code and yours though, so don’t know what’s happening :(

Nope sorry but you can find it online, just write Strassen algorithm C or C++

The function name should be strassen not strassen2, I just forgot to remove the 2 my bad

You're welcome!

Fixed in tests due to numpy. ----------------------- import unittest from ..strassen_with_numpy import strassen_with_numpy import numpy as np class MyTestCase(unittest.TestCase): def test_strassen(self): matrix_a = np.array([ [3, 5, 1, 3], [1, 2, 3, 4], [4, 5, 6, 8], [7, 8, 9, 3] ]) matrix_b = np.array([ [4, 1, 2, 3], [1, 2, 1, 6], [2, 4, 6, 2], [6, 2, 5, 4] ]) matrix_result = np.array([ [37, 23, 32, 53], [36, 25, 42, 37], [81, 54, 89, 86], [72, 65, 91, 99] ]) # stackoverflow.com/questions/3302949/best-way-to-assert-for-numpy-array-equality np.testing.assert_array_equal(strassen_with_numpy(matrix_a, matrix_b), matrix_result) if __name__ == '__main__': unittest.main()

Hello, yes we use numpy not Python lists

How did it go? Did you succeed?

@@insidecode Haha, so I actually switched my projects to a Vertex Cover and Convex Hull algorithms. I thought they'd be easier and more interesting to present than this would be.

@@insidecode Sorry to disappoint XD

No problem, by the way I made 2 convex hull videos, you can check them

@@insidecode Oh yeah, turns out those were the first ones I watched when looking at the algorithm a few weeks ago. I should link my results when I'm done.

Thanks a lot!