How to solve recursion in Fast Min Cut algorithm?

Question

I'm trying to implement a code of improvement of Karger's algorithm for finding a min-cut in a graph. I've an array of vertices and a matrix M, where M_ij are numbers of edges betwen vertices i and j. All algorithm uses a double recursion, if number of vertices isn't smaller than 6 vertices. But how I should solve it?

cuts=[]

def rekursion(M, vertices, n):
    global cuts
    t = len(vertices)
    if t<=6:
        n, vertices0, M0 = min_cut_part(M, vertices, 2, n) #makes Karger's algorithm till 2 vertices remains
        cuts.append(int(max(map(max, M0)))) #adds the only none-zero element of matrix, which is actually a cut
        return cuts
    else:
        t = int(1+t/np.sqrt(2))
        n, vertices1, M1 = min_cut_part(M, vertices, t, n)
        n, vertices2, M2 = min_cut_part(M, vertices, t, n)
        rekursion(M1, vrcholy1, n)
        rekursion(M2, vrcholy2, n)
        
def fast_min_cut(edges):
    n, verices, M = matrix_vertices(edges) #creates a matrix representing edges
    rekursion(M, vertices, n)
    print("cuts:",cuts)
    print("min-cut:",min(cuts))

I already found out, that I must use global cuts. But still if I run a code, the result is as follows:result Probably the recursion always use the same inter-graph..

Please, would you help me, how to solve it?