Pymatching Toric Code vs Surface Code

Question

I am looking the example of toric code in Pymatching.

Here is the code:

 import numpy as np
import matplotlib.pyplot as plt
from scipy.sparse import hstack, kron, eye, csr_matrix, block_diag


def repetition_code(n):
    
    row_ind, col_ind = zip(*((i, j) for i in range(n) for j in (i, (i+1)%n)))
    data = np.ones(2*n, dtype=np.uint8)
    return csr_matrix((data, (row_ind, col_ind)))


def toric_code_x_stabilisers(L):
    Hr = repetition_code(L)
    H = hstack(
            [kron(Hr, eye(Hr.shape[1])), kron(eye(Hr.shape[0]), Hr.T)],
            dtype=np.uint8
        )
    H.data = H.data % 2
    H.eliminate_zeros()
    return csr_matrix(H)
    
    
def toric_code_x_logicals(L):
   
    H1 = csr_matrix(([1], ([0],[0])), shape=(1,L), dtype=np.uint8)
    H0 = csr_matrix(np.ones((1, L), dtype=np.uint8))
    x_logicals = block_diag([kron(H1, H0), kron(H0, H1)])
    x_logicals.data = x_logicals.data % 2
    x_logicals.eliminate_zeros()
    return csr_matrix(x_logicals)

1. I want to test minumum weight perfect matching in Pymatching. I want to run this code in the absence of noise and I want to see what will be the result of pymatching. To do that, I use the following :

   H=toric_code_x_stabilisers(4)

   matching = pymatching.Matching(H)

   matching.draw()

Is that correct? I see nothing, is it because I have no error in the system or is it because there is some issue with matching.draw() function? Then I tried the following code scripts:

 H=toric_code_x_stabilisers(4)
noise = np.array([0,0,0,0]) #no error
z = H@noise % 2
m=Matching(H)
print(m)
m.draw()
c = m.decode(z)

And I got the dimension mismatch error..

2. I want to simulate surface code in the absence of error. It is simply I will initialize my data qubits in zero states and I will use x and z stabilizer. I will measure the stabilizers and I will see the result. For example, when I measure the X stabilizer, they will commute with Z stabilizer etc.. In that case, I am not sure what should change in toric code to make it surface code. It seemed to me that, toric_code_x_stabilisers can stay same. Do we have a small example for surface code?

Answer 1

Here is the working code example:

H=toric_code_x_stabilisers(3)
noise = np.array([0,0,0,0]) #no error
z = H@noise % 2
m=Matching(H)
print(m)
m.draw()
#c = m.decode(z)#since there is no error, no decoding too. I simply closed the line