Dynamic graph in matplotlib with multiple graphs in one figure

Question

I've managed to set up dynamic graph of one variable in matplotlib:

def update_line(hl, new_data):

    hl.set_xdata(np.append(hl.get_xdata(), new_data[0]))
    hl.set_ydata(np.append(hl.get_ydata(), new_data[1]))
    plt.draw()


cost_plot, = plt.plot([], [], 'b-')
plt.xlabel('iter')
plt.ylabel('cost')
plt.axis([0, set_size, 0, 10])

some for:
    ...
    update_line(cost_plot, [iter, cost])
    plt.draw()
    plt.pause(0.001)

And with this code I'm plotting my cost function of my first neural network:

Graph of cost fun (network's still dead)

Now I want to do the same but for set of synapses, so I want multiple plots in one figure, updated dynamically by

update_line(plot, [iter, set of next values])

but I can't really find a way of doing this.

My data is stored in numpy's array

Final result would look something like this

Edit: Final result after doing it correctly

Have multiple calls to update_line and pass in the correct line and data? — DavidG
– DavidG, Commented May 3, 2018 at 10:23
plot is line data object returned by plt.plot - it's cost_plot in working example. Set of next values is an array of values I want to plot, set of next values[0] - next value of first plot, set of next values[1], next value of second plot and so on. — Konrad Rzońca
– Konrad Rzońca, Commented May 3, 2018 at 10:50
It works with DavidG's advice, I will post solution in a moment — Konrad Rzońca
– Konrad Rzońca, Commented May 3, 2018 at 10:56

Kefeng91 · Accepted Answer · 2018-05-03 10:47:58Z

1

You probably want something like this :

import numpy as np
import matplotlib.pyplot as plt

def update_line(hl, i, new_data):
    hl.set_data(np.arange(i+1), new_data[:i+1])

n_data = 2
n_iter = 10
data = np.random.rand(n_data, n_iter)

plt.figure()
plt.xlabel("iter")
plt.ylabel("cost")
plt.axis([0, n_iter, 0, 1])

cost_plots = []
for i in range(n_data):
    cost_plot, = plt.plot([], [])
    cost_plots.append(cost_plot)

for i in range(n_iter):
    for j, cost_plot in enumerate(cost_plots):
        update_line(cost_plot, i, data[j])
    plt.draw()
    plt.pause(0.1)

I created a list of cost_plots and pass it to your function update_line. At each iteration, I loop over and update the different plots, and update the figure at the end of the iteration.

answered May 3, 2018 at 10:47

Kefeng91

8126 silver badges10 bronze badges

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

1 Comment

Konrad Rzońca Over a year ago

Yeah, that's almost exactly what I did just now and it's correct solution. I didn't understand what exactly plt.plot returns and I wanted somehow to place all my plot data in one Line2D object. When you make a list just like you did you can update everyone correctly

Collectives™ on Stack Overflow

Dynamic graph in matplotlib with multiple graphs in one figure

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related