Vfrdtyky

I have a number of data series which I want to save as a video. The different data series should appear one by one, keeping the former series. In the end, all the series is shown in the plot. I want to do this with multiple data series/subplots in a synchronized way. I want the subplots to have different sizes, so - as far as I have understood - I need to do this with GridSpec. However, with GridSpec I am not able to keep the old series while adding new ones (I only get the last data series).

An example of my code is (showing the principle - my actual script is more complex):
Working with index/pos (but only equal size on subplots):

import numpy as np

import matplotlib.pyplot as plt

import matplotlib.animation as anim



figure = plt.figure()



data = np.array([1,2,3,4,5])

set = np.array([1,2,3,4])



def make_frame(i):

    ax1 = plt.subplot(121)

    x = data

    y = i*x

    ax1.set_ylim(0,25)

    ax1.plot(x,y)



    ax2 = plt.subplot((122), sharey = ax1)

    a = data*2

    b = i/x*3

    ax2.plot(a,b) 



    plt.pause(1)



ani = anim.FuncAnimation(figure, make_frame, frames = set, repeat = False)



plt.show()

Not keeping old series (but with ability of adjust number of columns/rows for each plot):

import numpy as np

import matplotlib.pyplot as plt

import matplotlib.animation as anim

from matplotlib.gridspec import GridSpec





figure = plt.figure()



data = np.array([1,2,3,4,5])

set = np.array([1,2,3,4])



print(data,set)



def make_frame(i):

    gs1 = GridSpec(1,2)



    ax1 = plt.subplot(gs1[0,0])

    x = data

    y = i*x

    ax1.plot(x,y)



    ax2 = plt.subplot(gs1[0,1], sharey = ax1)

    a = data

    b = i/x*3

    ax2.plot(a,b)



    ax1.set_ylim(0,25)



    plt.pause(0.5)



ani = anim.FuncAnimation(figure, make_frame, frames = set, repeat = False)



plt.show()

How can I keep the "plotting history" in the GridSpec solution?

You need to define the gridspec and the subplots outside of the animating function. Else they are recreated from scratch for each frame.

Also, do not use plt.pause in an animation, but instead use the interval argument. And don't use the names of python functions (set) as variable names.

import numpy as np

import matplotlib.pyplot as plt

import matplotlib.animation as anim

from matplotlib.gridspec import GridSpec



data = np.array([1,2,3,4,5])

seti = np.array([1,2,3,4])





gs1 = GridSpec(1,2)

figure = plt.figure()

ax1 = plt.subplot(gs1[0,0])

ax2 = plt.subplot(gs1[0,1], sharey = ax1)

ax1.set_ylim(0,25)



print(data,seti)



def make_frame(i):

    x = data

    y = i*x

    ax1.plot(x,y)



    a = data

    b = 3.*i/x

    ax2.plot(a,b)



ani = anim.FuncAnimation(figure, make_frame, frames = seti, 

                         interval =500, repeat = False)



plt.show()