animate_arrays

This module wraps some setup scripts and functions to make Matplotlib’s animations a little simpler to interface with. The starting point here becomes creating data with

Install

pip install animate_arrays

How to use 2D arrays

Import libraries

The animate_arrays functions work in Jupyter notebooks. You’ll need to import some of the libraries to build some arrays, plots, and animations as such,

import numpy as np
import matplotlib.pyplot as plt
from matplotlib import animation
from IPython.display import HTML

I prefer to use the fivethirtyeight style sheet, but you can use any style.

plt.style.use('fivethirtyeight')

Then, import the animate_lines function from the animate_arrays.animate module.

from animate_arrays.animate import animate_lines

Build arrays and watch them move

With the libraries imported, I build 2 arrays, X and Y where each column is $\times 3~(x,~y)$ coordinates that define two rotating arms.

• Arm 1 rotates once around the origin
• Arm 2 is connected to arm 1 and rotates twice

a = np.linspace(0, 2*np.pi, 100)

x1 = np.cos(a)
y1 = np.sin(a)

x2 = np.cos(2*a)
y2 = np.sin(2*a)
X = np.array([np.zeros(len(a)), x1, x1+x2])
Y = np.array([np.zeros(len(a)), y1, y1+y2])

a = animate_lines(X, Y)
HTML(a.to_html5_video())

The animation does not have equal axes, so the arms appear to be changing lengths as the rotations occur.

The setup_fig_function option allows you to define a custom plot setup. Here, I name it fig_setup

def fig_setup():
    fig, ax = plt.subplots()
    
    ax.plot(X[2, :], Y[2, :], '--')
    ax.axis('equal')
    return fig, ax

The setup_fig_function should not take any arguments. I used it to set up 2 new display additions, 1. plot the path of arm 2’s end so we can see where its been and where its going on the dashed line 2. equal axes with ax.axis('equal')

Below, I try running the function and looking at the resulting static figure for the animation

fig, ax = fig_setup()

a = animate_lines(X, Y, setup_fig_function= fig_setup)

HTML(a.to_html5_video())

How to use 3D arrays

In the first example, we used a 2D array where each column defined the line for a point in time. The animate_lines function can also plot multiple lines in each frame. Here, we’ll add another set of rotating arms.

a = np.linspace(0, 2*np.pi, 100)

x1 = np.cos(a)
y1 = np.sin(a)

x2 = np.cos(2*a)
y2 = np.sin(2*a)

x3 = np.cos(2*a)
y3 = np.sin(2*a)

x4 = np.cos(3*a)
y4 = np.sin(3*a)

X = np.zeros((3, 2, len(a)))
Y = np.zeros((3, 2, len(a)))

X[:, 0, :] = np.array([np.zeros(len(a)), x1, x1+x2])
X[:, 1, :] = np.array([np.zeros(len(a)), x3, x3+x4])

Y[:, 0, :] = np.array([np.zeros(len(a)), y1, y1+y2])
Y[:, 1, :] = np.array([np.zeros(len(a)), y3, y3+y4])

def fig_setup():
    fig, ax = plt.subplots()
    
    ax.plot(X[2, 0, :], Y[2, 0, :], '--')
    ax.plot(X[2, 1, :], Y[2, 1, :], '--')
    ax.axis('equal')
    return fig, ax

fig_setup()

(<Figure size 640x480 with 1 Axes>, <AxesSubplot: >)

a2 = animate_lines(X, Y, setup_fig_function=fig_setup)
HTML(a2.to_html5_video())