Live Graphics Using Matplotlib Not Showing the Plot

Question

I am working on a physics problem and trying to watch the simulation "live" by plotting after every 4 steps of the calculation. I am using Runge-Kutta (4th Order) method for the differentiation if that is relevant.

When I run the code (which should run the code for approximately 10/4 seconds, but I want to speed this up, though I am not sure how besides showing only some of the frames), I can see the axes moving, but nothing is showing in the plot.

I have stripped my code down to just generating random numbers instead of the actual function so I can share code as close to what I actually have.

import numpy as np
import matplotlib.pyplot as plt
 
# Parameters
time = 10  # maximum time for the simulation
h = 0.1  # step size
steps = int(time/h)  # number of steps
order = 4  # two second order equations
 
ICs = [0, 0, 0, 0.25, 0]  # intial conditions; t0, x1, x1dot, x2, x2dot
 
# Initializing vars array
vars = np.empty(shape=(order,steps))  # each row is another var, i.e. x1,x2,...
 
# Set initial conditions for each var
for i in range(order):
    vars[i][0] = ICs[i+1]

t = np.empty(steps)
t[0] = ICs[0]

K = np.zeros(shape=(4,len(vars))) # Each row is k1, k2, k3, k4 for each var


# ODE function
def ODE():
    dx1dt = np.random.randint(-1000,1000)
    dv1dt = np.random.randint(-1000,1000)
    dx2dt = np.random.randint(-1000,1000)
    dv2dt = np.random.randint(-1000,1000)
    
    return(np.array([dx1dt, dv1dt, dx2dt, dv2dt]))

# Loop calculates each var value using RK 4th order method
for i in range(steps-1):
    K[0] = ODE()
    K[1] = ODE()
    K[2] = ODE()
    K[3] = ODE()
    
    vars[:,i+1] = vars[:,i] + 1/6 * (K[0] + 2 * K[1] + 2 * K[2] + K[3])
    
    t[i+1] = t[i] + h
    
    # Plotting every fourth calculation
    if (i%4 == 0):
        plt.cla()
        plt.plot(t[i], vars[0,i], label='x1')
        
        plt.title(f'Title (stepsize: {h})')
        plt.xlabel('time [s]')
        plt.ylabel('position [m]')
        
        plt.legend(loc=1)
        plt.tight_layout()
        plt.pause(0.01)
 
plt.tight_layout()
 
plt.show()

Thanks for the help.

at a first glance you're plotting a single point at each iteration, maybe you could plot the arrays instead? — filippo
– filippo, Commented Jun 22, 2020 at 18:02
When do plt.plot(t, vars[0], label='x1') instead, there seems to be some initial plot despite there being no values yet. If I initialized vars as vars = np.zeros(shape=(order,steps)), I see the values slowly updating, but I want to instead see graph developing, if that makes sense. — jared
– jared, Commented Jun 22, 2020 at 19:43

Sameeresque · Accepted Answer · 2020-06-22 22:19:02Z

1

You could make use of matplotlib's interactive mode plt.ion() to initiate interactive plotting. Then call plt.show() to display the window, and update it using plt.gcf().canvas.draw, as shown here:

import numpy as np
import matplotlib.pyplot as plt
  
# Parameters
time = 10  # maximum time for the simulation
h = 0.1  # step size
steps = int(time/h)  # number of steps
order = 4  # two second order equations
 
ICs = [0, 0, 0, 0.25, 0]  # intial conditions; t0, x1, x1dot, x2, x2dot
 
# Initializing vars array
vars = np.empty(shape=(order,steps))  # each row is another var, i.e. x1,x2,...
 
# Set initial conditions for each var
for i in range(order):
    vars[i][0] = ICs[i+1]

t = np.empty(steps)
t[0] = ICs[0]

K = np.zeros(shape=(4,len(vars))) # Each row is k1, k2, k3, k4 for each var
fig,ax=plt.subplots()
plt.ion() # set interactive mode on 
plt.show() # open display window

# ODE function
def ODE():
    dx1dt = np.random.randint(-1000,1000)
    dv1dt = np.random.randint(-1000,1000)
    dx2dt = np.random.randint(-1000,1000)
    dv2dt = np.random.randint(-1000,1000)
    
    return(np.array([dx1dt, dv1dt, dx2dt, dv2dt]))

# Loop calculates each var value using RK 4th order method
for i in range(steps-1):
    K[0] = ODE()
    K[1] = ODE()
    K[2] = ODE()
    K[3] = ODE()
    
    vars[:,i+1] = vars[:,i] + 1/6 * (K[0] + 2 * K[1] + 2 * K[2] + K[3])
    
    t[i+1] = t[i] + h
    
    # Plotting every fourth calculation
    if (i%4 == 0):
        #plt.cla()
        plt.plot(t[:i], vars[0,:i],color='black')
        
        plt.title(f'Title (stepsize: {h})')
        plt.xlabel('time [s]')
        plt.ylabel('position [m]')
        

        plt.tight_layout()
        plt.gcf().canvas.draw() #update display window
        plt.pause(0.01)
        
plt.tight_layout()

edited Jun 22, 2020 at 22:19

answered Jun 22, 2020 at 20:56

Sameeresque

2,6022 gold badges14 silver badges24 bronze badges

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

3 Comments

jared Over a year ago

This is great. I guess I have to look into this interactive mode now. Thanks!

jared Over a year ago

How did you save this gif? I always have a hard time creating a gif of a graph...

Sameeresque Over a year ago

Save them as individual frames and stack them to create a gif. You will find this SO helpful.

milanovitch44 · Accepted Answer · 2020-06-22 17:53:08Z

0

It might not work because your plt.show() is not in the in the if clause.

answered Jun 22, 2020 at 17:53

milanovitch44

214 bronze badges

1 Comment

jared Over a year ago

Adding plt.show() to the if statement appears to have no effect.

Collectives™ on Stack Overflow

Live Graphics Using Matplotlib Not Showing the Plot

2 Answers 2

3 Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related