Python: Update plot instead of creating a new one

Question

I am trying to code the Ising Model using Python.

I think, I've coded it correctly, but I have a problem with the animation or the plotting. I seem to plot a new image of every configuration, instead of updating the existing one, resulting in a lot of saved images that I don't need. I just want a single plot that is being updated, if that's possible.

I know, I am plotting inside the loop, but I don't recall that being an issue, when I want to plot every iteration. Can it be a problem with Seaborn's heatmap?

I've attached my code:

import numpy as np
import numpy.random as npr
import matplotlib.pyplot as plt
import seaborn as sns

#Constants
J = 1
h = 1
kbT = 1
beta = 1

#Grid
L = 20 #Dimensions
N = L**2 #Total number of grid points


#Initial configuration
spins = 2*np.random.randint(2, size = (L,L))-1


E = []
i = 0
plt.figure()

while i < 100000:
    for i in range(1,N):
        i += 1
        s = tuple(npr.randint(0, L, 2)) # Random initial coordinate
    
        # x and y coordinate
        (sx, sy) = s
        # Periodic boundary condition
        sl = (sx-1, sy) 
        sr = ((sx+1)%L, sy)
        sb = (sx, sy-1)
        st = (sx, (sy+1)%L)
        # Energy
        E =   spins[s] * ( spins[sl] + spins[sr] + spins[sb] + spins[st] )
        if E <= 0 : # If negative, flip
            spins[s] *= -1
        else:
            x = np.exp(-E/kbT) # If positve, check condition
            q = npr.rand()
            if x > q: 
                spins[s] *= -1
    # Plot (heatmap)
    sns.heatmap(spins, cmap = 'magma')
    plt.pause(10e-10)
    plt.draw()
    plt.show()

I think you are looking for the function ion, you can use it as plt.ion(), take a look here — Carlo Zanocco
– Carlo Zanocco, Commented Sep 5, 2020 at 12:06

Carlo Zanocco · Accepted Answer · 2020-09-05 12:20:33Z

1

I think the function ion and clf could do the trick.

import numpy as np
import numpy.random as npr
import matplotlib.pyplot as plt
import seaborn as sns

#Constants
J = 1
h = 1
kbT = 1
beta = 1

#Grid
L = 20 #Dimensions
N = L**2 #Total number of grid points

#Initial configuration
spins = 2*np.random.randint(2, size = (L,L))-1

E = []
i = 0

plt.ion()
plt.figure()
plt.show()

while i < 100000:
    for i in range(1,N):
        i += 1
        s = tuple(npr.randint(0, L, 2)) # Random initial coordinate
    
        # x and y coordinate
        (sx, sy) = s
        # Periodic boundary condition
        sl = (sx-1, sy) 
        sr = ((sx+1)%L, sy)
        sb = (sx, sy-1)
        st = (sx, (sy+1)%L)
        # Energy
        E =   spins[s] * ( spins[sl] + spins[sr] + spins[sb] + spins[st] )
        if E <= 0 : # If negative, flip
            spins[s] *= -1
        else:
            x = np.exp(-E/kbT) # If positve, check condition
            q = npr.rand()
            if x > q: 
                spins[s] *= -1
    # Plot (heatmap)
    plt.clf()
    sns.heatmap(spins, cmap = 'magma')
    plt.pause(10e-10)

With the function ion you are making interactive the plot, so you need to:

Make it interactive
Show the plot
Clear the plot in your cycle

Here the reference for the ion function.

Reference for clf is here

answered Sep 5, 2020 at 12:20

Carlo Zanocco

2,0294 gold badges21 silver badges33 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

JohnGoodWill Over a year ago

That did the trick! Thank you. Now I only get one "<Figure size XxY with 0 Axes>", instead of a lot.

Collectives™ on Stack Overflow

Python: Update plot instead of creating a new one

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related