How to use the thewalrus._torontonian.tor function in thewalrus

prev_prob = 1.0
    mu = np.zeros(n1)

    for k in range(nmodes):
        probs1 = np.zeros([2], dtype=np.float64)
        kk = np.arange(k + 1)
        _, V_red = reduced_gaussian(mu, cov, kk)

        Q = Qmat(V_red, hbar=hbar)
        A = Amat(Q, hbar=hbar, cov_is_qmat=True)
        O = Xmat(k + 1) @ A

        indices = result + [0]
        ind2 = indices + indices

        probs1[0] = tor(np.complex128(reduction(O, ind2))).real

        indices = result + [1]
        ind2 = indices + indices
        pref = np.sqrt(np.linalg.det(Q).real)
        probs1a = probs1 / pref

        probs2 = probs1a / prev_prob
        probs2[1] = 1.0 - probs2[0]
        probs1a[1] = probs2[1] * prev_prob
        probs3 = np.maximum(
            probs2, np.zeros_like(probs2)
        )  # pylint: disable=assignment-from-no-return
        probs3 /= np.sum(probs3)
        result.append(np.random.choice(a=range(len(probs3)), p=probs3))

        prev_prob = probs1a[result[-1]]

prev_prob = 1.0
    mu = np.zeros(n1)

    for k in range(nmodes):
        probs1 = np.zeros([2], dtype=np.float64)
        kk = np.arange(k + 1)
        _, V_red = reduced_gaussian(mu, cov, kk)

        Q = Qmat(V_red, hbar=hbar)
        A = Amat(Q, hbar=hbar, cov_is_qmat=True)
        O = Xmat(k + 1) @ A

        indices = result + [0]
        ind2 = indices + indices

        probs1[0] = tor(np.complex128(reduction(O, ind2))).real

        indices = result + [1]
        ind2 = indices + indices
        pref = np.sqrt(np.linalg.det(Q).real)
        probs1a = probs1 / pref

        probs2 = probs1a / prev_prob
        probs2[1] = 1.0 - probs2[0]
        probs1a[1] = probs2[1] * prev_prob
        probs3 = np.maximum(
            probs2, np.zeros_like(probs2)
        )  # pylint: disable=assignment-from-no-return
        probs3 /= np.sum(probs3)
        result.append(np.random.choice(a=range(len(probs3)), p=probs3))

        prev_prob = probs1a[result[-1]]