plotnine.animation.PlotnineAnimation

class plotnine.animation.PlotnineAnimation(plots, interval=200, repeat_delay=None, repeat=True, blit=False)[source]

Animation using ggplot objects

Parameters:
plots : iterable

ggplot objects that make up the the frames of the animation

interval : number, optional

Delay between frames in milliseconds. Defaults to 200.

repeat_delay : number, optional

If the animation in repeated, adds a delay in milliseconds before repeating the animation. Defaults to None.

repeat : bool, optional

Controls whether the animation should repeat when the sequence of frames is completed. Defaults to True.

blit : bool, optional

Controls whether blitting is used to optimize drawing. Defaults to False.

Notes

  1. The plots should have the same facet and the facet should not have fixed x and y scales.
  2. The scales of all the plots should have the same limits. It is a good idea to create a scale (with limits) for each aesthetic and add them to all the plots.
  3. For plots with legends or any other features that are cutoff, use the subplots_adjust themeable to create space for it.

Examples

In [1]:
import pandas as pd
import numpy as np
from plotnine import *
from plotnine.animation import PlotnineAnimation

# for animation in the notebook
from matplotlib import rc
rc('animation', html='html5')

Spiral Animation

How to create Animations

The PlotnineAnimation methods takes a sequence of plots and animates them. The best way to create the sequence is to use a function that generates plots.

NOTE: When creating the plots make sure all the plots have scales with the same limits for each aesthetic, including the x and y aesthetics.

In [2]:
# Parameters used to control the spiral
n = 100
tightness = 1.3
kmin = 1
kmax = 25
num_frames = 25
theta = np.linspace(-np.pi, np.pi, n)

def plot(k):
    # For every plot we change the theta
    _theta = theta*k

    # Polar Equation of each spiral
    r = tightness*_theta

    df = pd.DataFrame({
        'theta': _theta,
        'r': r,
        'x': r*np.sin(_theta),
        'y': r*np.cos(_theta)
    })

    p = (ggplot(df)
         + geom_path(aes('x', 'y', color='theta'), size=1)
         + lims(
             # All the plots have scales with the same limits
             x=(-130, 130),
             y=(-130, 130),
             color=(-kmax*np.pi, kmax*np.pi)
         )
         + theme_void()
         + theme(
             aspect_ratio=1,
             # Make room on the right for the legend
             subplots_adjust={'right': 0.85}
         )
    )
    return p

# It is better to use a generator instead of a list
plots = (plot(k) for k in np.linspace(kmin, kmax, num_frames))
ani = PlotnineAnimation(plots, interval=100, repeat_delay=500)
# ani.save('/tmp/animation.mp4')
ani
Out[2]:

The final image of the spiral

In [3]:
plot(kmax)
../_images/PlotnineAnimatio_5_0.png
../_images/PlotnineAnimatio_5_1.png
Out[3]:
<ggplot: (97654321012345679)>