Hi Kwant Users

I notice an inconsistency in calculation result using Python IDLE (3.5.1, Anaconda3).

The issue is that every time I execute the same code, it gives me a different result.The test code that I use is the example code for graphene quantum dot calculation (2.7.1) .

First, I used the different parameter (r=3). Each time I executed the same code, it returned different wavefunction. If I use a large r (r=8), the wavefunction(n=225) is consistent. But if I plot the wavefunction 22 instead of 225 as an example, then the wavefunction is different for each time.  I tested the code within jupyter notebook and it doesn't show any inconsistencies. 

Can anyone advise me what raises this issue ? Is it because of Kwant package or Python IDLE or the calculation script itself?

If I still want to use Python IDLE to do the calculation. How can I make sure my calculation result is correct?


The test code is copied at the end of the email.

Thanks in advance.

Sincerely,
Liming

Test code:
================================================================
================================================================

# Tutorial 2.7.1. 2D example: graphene quantum dot
# ================================================
#
# Physics background
# ------------------
#  - graphene edge states
#
# Kwant features highlighted
# --------------------------
#  - demonstrate different ways of plotting

import kwant
from matplotlib import pyplot

lat = kwant.lattice.honeycomb()
a, b = lat.sublattices

def make_system(r=8, t=-1, tp=-0.1):

    def circle(pos):
        x, y = pos
        return x**2 + y**2 < r**2

    sys = kwant.Builder()
    sys[lat.shape(circle, (0, 0))] = 0
    sys[lat.neighbors()] = t
    sys.eradicate_dangling()
    if tp:
        sys[lat.neighbors(2)] = tp

    return sys


def plot_system(sys):
    kwant.plot(sys)
    # the standard plot is ok, but not very intelligible. One can do
    # better by playing wioth colors and linewidths

    # use color and linewidths to get a better plot
    def family_color(site):
        return 'black' if site.family == a else 'white'

    def hopping_lw(site1, site2):
        return 0.04 if site1.family == site2.family else 0.1

    kwant.plot(sys, site_lw=0.1, site_color=family_color, hop_lw=hopping_lw)


def plot_data(sys, n):
    import scipy.linalg as la

    sys = sys.finalized()
    ham = sys.hamiltonian_submatrix()
    evecs = la.eigh(ham)[1]

    wf = abs(evecs[:, n])**2

    # the usual - works great in general, looks just a bit crufty for
    # small systems

    kwant.plotter.map(sys, wf, oversampling=10, cmap='gist_heat_r')

    # use two different sort of triangles to cleanly fill the space
    def family_shape(i):
        site = sys.sites[i]
        return ('p', 3, 180) if site.family == a else ('p', 3, 0)

    def family_color(i):
        return 'black' if sys.site(i).family == a else 'white'

    kwant.plot(sys, site_color=wf, site_symbol=family_shape,
               site_size=0.5, hop_lw=0, cmap='gist_heat_r')

    # plot by changing the symbols itself
    def site_size(i):
        return 3 * wf[i] / wf.max()

    kwant.plot(sys, site_size=site_size, site_color=(0, 0, 1, 0.3),
               hop_lw=0.1)


def main():
    # plot the graphene system in different styles
    sys = make_system()

    plot_system(sys)

    # compute a wavefunction (number 225) and plot it in different
    # styles
    sys = make_system(tp=0)

    plot_data(sys, 22)


# Call the main function if the script gets executed (as opposed to imported).
# See <http://docs.python.org/library/__main__.html>.
if __name__ == '__main__':
    main()