Hi,
I've been "hard" at work over the last couple of months putting
together a set of classes that simplifies the creation of animations
in matplotlib. This started when I resurrected some old code for
animations to give to a colleague, when I realized just how bad the
old code was and how much better I could do. The result of this
"afternoon" hack is what I'm ready to put forth. Some of the goals:
* Run independently of the backend, unlike the examples we have now
(This is really accomplished by the Timer object we now have)
* Remove the boilerplate code of setting up loops
* Facilitate saving out animations as a movie file
* Provide a simple API that integrates well with the rest of Matplotlib
* Provide (optional) blitting support so that users don't have to
learn the ins and outs of blitting
Overall, I think I accomplished my goals, so I'm putting this out
there for wider comments. I've attached the python module which, when
run, displays two animated figures. There is also a git repository at:
http://github.com/dopplershift/Animation
which has some more examples, including ports of our old examples.
(The examples assume animation.py is in your python path somewhere,
which you'll have to do by hand. This can be as simple as dropping
animation.py into the directory).
Some things to note:
* The flow is broken into *a lot* of member functions. This is to
provide sufficient entry points for subclasses so that they really
only need to reimplement the parts they override. Optional blitting
support drove a lot of this.
* There are two main classes for end users:
* FuncAnimation -- provide a callback which draws the next frame of animation
* ArtistAnimation -- provide a sequence of collections of artists
which are turned off and on for each frame of animation
* There is support for saving movies with either mencoder or ffmpeg.
The config for this is really rough, and the place I could *really*
use suggestions. I'm not sure how best to go about it. I've been
unable to find a (currently maintained) python library for saving
movie files, so system calls to the utilities is the best I can do at
the moment. I'm not sure what to use on windows, since I'm not sure of
the state (and requirements) of mencode/ffmpeg on windows.
TODOs:
* Configuring saving movie files (formats, programs, etc.) (see above)
* Documentation (I promise not to commit until this is written)
* More examples (could use some more procedural examples, like
animating using data read from a socket, or inotify)
I welcome feedback on this, because I really want to see this become
an easy and bulletproof way of doing animations in matplotlib. This
seems to be an area of frequent question on the mailing list, and I
want this framework to lessen the questions, not increase them.
Ryan
--
Ryan May
Graduate Research Assistant
School of Meteorology
University of Oklahoma
# TODO:
# * Documentation -- this will need a new section of the User's Guide.
# Both for Animations and just timers.
# - Also need to update http://www.scipy.org/Cookbook/Matplotlib/Animations
# * Blit
# * Currently broken with Qt4 for widgets that don't start on screen
# * Still a few edge cases that aren't working correctly
# * Can this integrate better with existing matplotlib animation artist flag?
# * Example
# * Frameless animation - pure procedural with no loop
# * Need example that uses something like inotify or subprocess
# * Complex syncing examples
# * Movies
# * Library to make movies?
# * RC parameter for config?
# * Need to consider event sources to allow clicking through multiple figures
from datetime import datetime
def traceme(func):
def wrapper(*args):
print '%s -- Calling: %s %s' % (datetime.now(), func.__name__, str(args))
ret = func(*args)
print 'Returned: %s' % func.__name__
return ret
return wrapper
from matplotlib.cbook import iterable
class Animation(object):
'''
This class wraps the creation of an animation using matplotlib. It is
only a base class which should be subclassed to provide needed behavior.
*fig* is the figure object that is used to get draw, resize, and any
other needed events.
*event_source* is a class that can run a callback when desired events
are generated, as well as be stopped and started. Examples include timers
(see :class:`TimedAnimation`) and file system notifications.
*blit* is a boolean that controls whether blitting is used to optimize
drawing.
'''
def __init__(self, fig, event_source=None, blit=False):
self._fig = fig
self._blit = blit
# These are the basics of the animation. The frame sequence represents
# information for each frame of the animation and depends on how the
# drawing is handled by the subclasses. The event source fires events
# that cause the frame sequence to be iterated.
self.frame_seq = self.new_frame_seq()
self.event_source = event_source
# Clear the initial frame
self._init_draw()
# Instead of starting the event source now, we connect to the figure's
# draw_event, so that we only start once the figure has been drawn.
self._first_draw_id = fig.canvas.mpl_connect('draw_event', self._start)
# Connect to the figure's close_event so that we don't continue to
# fire events and try to draw to a deleted figure.
self._close_id = self._fig.canvas.mpl_connect('close_event', self._stop)
if blit:
self._setup_blit()
def _start(self, *args):
'''
Starts interactive animation. Adds the draw frame command to the GUI
handler, calls show to start the event loop.
'''
# On start, we add our callback for stepping the animation and
# actually start the event_source. We also disconnect _start
# from the draw_events
self.event_source.add_callback(self._step)
self.event_source.start()
self._fig.canvas.mpl_disconnect(self._first_draw_id)
def _stop(self, *args):
# On stop we disconnect all of our events.
if self._blit:
self._fig.canvas.mpl_disconnect(self._resize_id)
self._fig.canvas.mpl_disconnect(self._close_id)
self.event_source.remove_callback(self._step)
self.event_source = None
def save(self, filename, fps=5, codec='mpeg4', clear_temp=True,
frame_prefix='_tmp'):
'''
Saves a movie file by drawing every frame.
*fps* is the frames per second in the movie
*codec* is the codec to be used,if it is supported by the output method.
*clear_temp* specifies whether the temporary image files should be
deleted.
*frame_prefix* gives the prefix that should be used for individual
image files. This prefix will have a frame number (i.e. 0001) appended
when saving individual frames.
'''
fnames = []
# Create a new sequence of frames for saved data. This is different
# from new_frame_seq() to give the ability to save 'live' generated
# frame information to be saved later.
for idx,data in enumerate(self.new_saved_frame_seq()):
self._draw_next_frame(data, blit=False)
fname = '%s%04d.png' % (frame_prefix, idx)
fnames.append(fname)
self._fig.savefig(fname)
self._make_movie(filename, fps, codec, frame_prefix)
#Delete temporary files
if clear_temp:
import os
for fname in fnames:
os.remove(fname)
def ffmpeg_cmd(self, fname, fps, codec, frame_prefix):
# Returns the command line parameters for subprocess to use
# ffmpeg to create a movie
return ['ffmpeg', '-y', '-r', str(fps), '-b', '1800k', '-i',
'%s%%04d.png' % frame_prefix, fname]
def mencoder_cmd(self, fname, fps, codec, frame_prefix):
# Returns the command line parameters for subprocess to use
# mencoder to create a movie
return ['mencoder', 'mf://%s*.png' % frame_prefix, '-mf',
'type=png:fps=%d' % fps, '-ovc', 'lavc', '-lavcopts',
'vcodec=%s' % codec, '-oac', 'copy', '-o', fname]
def _make_movie(self, fname, fps, codec, frame_prefix, cmd_gen=None):
# Uses subprocess to call the program for assembling frames into a
# movie file. *cmd_gen* is a callable that generates the sequence
# of command line arguments from a few configuration options.
from subprocess import Popen, PIPE
if cmd_gen is None:
cmd_gen = self.ffmpeg_cmd
proc = Popen(cmd_gen(fname, fps, codec, frame_prefix), shell=False,
stdout=PIPE, stderr=PIPE)
proc.wait()
def _step(self, *args):
'''
Handler for getting events. By default, gets the next frame in the
sequence and hands the data off to be drawn.
'''
# Returns True to indicate that the event source should continue to
# call _step, until the frame sequence reaches the end of iteration,
# at which point False will be returned.
try:
framedata = self.frame_seq.next()
self._draw_next_frame(framedata, self._blit)
return True
except StopIteration:
return False
def new_frame_seq(self):
'Creates a new sequence of frame information.'
# Default implementation is just an iterator over self._framedata
return iter(self._framedata)
def new_saved_frame_seq(self):
'Creates a new sequence of saved/cached frame information.'
# Default is the same as the regular frame sequence
return self.new_frame_seq()
def _draw_next_frame(self, framedata, blit):
# Breaks down the drawing of the next frame into steps of pre- and
# post- draw, as well as the drawing of the frame itself.
self._pre_draw(framedata, blit)
self._draw_frame(framedata)
self._post_draw(framedata, blit)
def _init_draw(self):
# Initial draw to clear the frame. Also used by the blitting code
# when a clean base is required.
pass
def _pre_draw(self, framedata, blit):
# Perform any cleaning or whatnot before the drawing of the frame.
# This default implementation allows blit to clear the frame.
if blit:
self._blit_clear(self._drawn_artists, self._blit_cache)
def _draw_frame(self, framedata):
# Performs actual drawing of the frame.
raise NotImplementedError('Needs to be implemented by subclasses to'
' actually make an animation.')
def _post_draw(self, framedata, blit):
# After the frame is rendered, this handles the actual flushing of
# the draw, which can be a direct draw_idle() or make use of the
# blitting.
if blit and self._drawn_artists:
self._blit_draw(self._drawn_artists, self._blit_cache)
else:
self._fig.canvas.draw_idle()
# The rest of the code in this class is to facilitate easy blitting
def _blit_draw(self, artists, bg_cache):
# Handles blitted drawing, which renders only the artists given instead
# of the entire figure.
updated_ax = []
for a in artists:
# If we haven't cached the background for this axes object, do
# so now. This might not always be reliable, but it's an attempt
# to automate the process.
if a.axes not in bg_cache:
bg_cache[a.axes] = a.figure.canvas.copy_from_bbox(a.axes.bbox)
a.axes.draw_artist(a)
updated_ax.append(a.axes)
# After rendering all the needed artists, blit each axes individually.
for ax in set(updated_ax):
ax.figure.canvas.blit(ax.bbox)
def _blit_clear(self, artists, bg_cache):
# Get a list of the axes that need clearing from the artists that
# have been drawn. Grab the appropriate saved background from the
# cache and restore.
axes = set(a.axes for a in artists)
for a in axes:
a.figure.canvas.restore_region(bg_cache[a])
def _setup_blit(self):
# Setting up the blit requires: a cache of the background for the
# axes
self._blit_cache = dict()
self._drawn_artists = []
self._resize_id = self._fig.canvas.mpl_connect('resize_event',
self._handle_resize)
self._post_draw(None, self._blit)
def _handle_resize(self, *args):
# On resize, we need to disable the resize event handling so we don't
# get too many events. Also stop the animation events, so that
# we're paused. Reset the cache and re-init. Set up an event handler
# to catch once the draw has actually taken place.
self._fig.canvas.mpl_disconnect(self._resize_id)
self.event_source.stop()
self._blit_cache.clear()
self._init_draw()
self._resize_id = self._fig.canvas.mpl_connect('draw_event', self._end_redraw)
def _end_redraw(self, evt):
# Now that the redraw has happened, do the post draw flushing and
# blit handling. Then re-enable all of the original events.
self._post_draw(None, self._blit)
self.event_source.start()
self._fig.canvas.mpl_disconnect(self._resize_id)
self._resize_id = self._fig.canvas.mpl_connect('resize_event',
self._handle_resize)
class TimedAnimation(Animation):
'''
:class:`Animation` subclass that supports time-based animation, drawing
a new frame every *interval* milliseconds.
*repeat* controls whether the animation should repeat when the sequence
of frames is completed.
*repeat_delay* optionally adds a delay in milliseconds before repeating
the animation.
'''
def __init__(self, fig, interval=200, repeat_delay=None, repeat=True,
event_source=None, *args, **kwargs):
# Store the timing information
self._interval = interval
self._repeat_delay = repeat_delay
self.repeat = repeat
# If we're not given an event source, create a new timer. This permits
# sharing timers between animation objects for syncing animations.
if event_source is None:
event_source = fig.canvas.new_timer()
event_source.interval = self._interval
Animation.__init__(self, fig, event_source=event_source, *args, **kwargs)
def _step(self, *args):
'''
Handler for getting events.
'''
# Extends the _step() method for the Animation class. If
# Animation._step signals that it reached the end and we want to repeat,
# we refresh the frame sequence and return True. If _repeat_delay is
# set, change the event_source's interval to our loop delay and set the
# callback to one which will then set the interval back.
still_going = Animation._step(self, *args)
if not still_going and self.repeat:
if self._repeat_delay:
self.event_source.remove_callback(self._step)
self.event_source.interval = self._repeat_delay
self.event_source.add_callback(self._loop_delay)
self.frame_seq = self.new_frame_seq()
return True
else:
return still_going
def _stop(self, *args):
# If we stop in the middle of a loop delay (which is relatively likely
# given the potential pause here, remove the loop_delay callback as
# well.
self.event_source.remove_callback(self._loop_delay)
Animation._stop(self)
def _loop_delay(self, *args):
# Reset the interval and change callbacks after the delay.
self.event_source.remove_callback(self._loop_delay)
self.event_source.interval = self._interval
self.event_source.add_callback(self._step)
class ArtistAnimation(TimedAnimation):
'''
Before calling this function, all plotting should have taken place
and the relevant artists saved.
frame_info is a list, with each list entry a collection of artists that
represent what needs to be enabled on each frame. These will be disabled
for other frames.
'''
def __init__(self, fig, artists, *args, **kwargs):
# Internal list of artists drawn in the most recent frame.
self._drawn_artists = []
# Use the list of artists as the framedata, which will be iterated
# over by the machinery.
self._framedata = artists
TimedAnimation.__init__(self, fig, *args, **kwargs)
def _init_draw(self):
# Make all the artists involved in *any* frame invisible
axes = []
for f in self.new_frame_seq():
for artist in f:
artist.set_visible(False)
# Assemble a list of unique axes that need flushing
if artist.axes not in axes:
axes.append(artist.axes)
# Flush the needed axes
for ax in axes:
ax.figure.canvas.draw()
def _pre_draw(self, framedata, blit):
'''
Clears artists from the last frame.
'''
if blit:
# Let blit handle clearing
self._blit_clear(self._drawn_artists, self._blit_cache)
else:
# Otherwise, make all the artists from the previous frame invisible
for artist in self._drawn_artists:
artist.set_visible(False)
def _draw_frame(self, artists):
# Save the artists that were passed in as framedata for the other
# steps (esp. blitting) to use.
self._drawn_artists = artists
# Make all the artists from the current frame visible
for artist in artists:
artist.set_visible(True)
class FuncAnimation(TimedAnimation):
'''
Makes an animation by repeatedly calling a function *func*, passing in
(optional) arguments in *fargs*.
*frames* can be a generator, an iterable, or a number of frames.
*init_func* is a function used to draw a clear frame. If not given, the
results of drawing from the first item in the frames sequence will be
used.
'''
def __init__(self, fig, func, frames=None ,init_func=None, fargs=None,
save_count=None, **kwargs):
if fargs:
self._args = fargs
else:
self._args = ()
self._func = func
# Amount of framedata to keep around for saving movies. This is only
# used if we don't know how many frames there will be: in the case
# of no generator or in the case of a callable.
self.save_count = save_count
# Set up a function that creates a new iterable when needed. If nothing
# is passed in for frames, just use itertools.count, which will just
# keep counting from 0. A callable passed in for frames is assumed to
# be a generator. An iterable will be used as is, and anything else
# will be treated as a number of frames.
if frames is None:
import itertools
self._iter_gen = itertools.count
elif callable(frames):
self._iter_gen = frames
elif iterable(frames):
self._iter_gen = lambda: iter(frames)
self.save_count = len(frames)
else:
self._iter_gen = lambda: iter(range(frames))
self.save_count = frames
# If we're passed in and using the default, set it to 100.
if self.save_count is None:
self.save_count = 100
self._init_func = init_func
self._save_seq = []
TimedAnimation.__init__(self, fig, **kwargs)
def new_frame_seq(self):
# Use the generating function to generate a new frame sequence
return self._iter_gen()
def new_saved_frame_seq(self):
# Generate an iterator for the sequence of saved data.
return iter(self._save_seq)
def _init_draw(self):
# Initialize the drawing either using the given init_func or by
# calling the draw function with the first item of the frame sequence.
# For blitting, the init_func should return a sequence of modified
# artists.
if self._init_func is None:
self._draw_frame(self.new_frame_seq().next())
else:
self._drawn_artists = self._init_func()
def _draw_frame(self, framedata):
# Save the data for potential saving of movies.
self._save_seq.append(framedata)
# Make sure to respect save_count (keep only the last save_count around)
self._save_seq = self._save_seq[-self.save_count:]
# Call the func with framedata and args. If blitting is desired,
# func needs to return a sequence of any artists that were modified.
self._drawn_artists = self._func(framedata, *self._args)
if __name__ == '__main__':
import numpy as np
import matplotlib.pyplot as plt
def update_line(num, data, line):
line.set_data(data[...,:num])
return line,
fig1 = plt.figure()
data = np.random.rand(2, 25)
l, = plt.plot([], [], 'r-')
plt.xlim(0, 1)
plt.ylim(0, 1)
plt.xlabel('x')
plt.title('test')
line_ani = FuncAnimation(fig1, update_line, 25, fargs=(data, l), interval=50, blit=True)
# line_ani.save('lines.mp4')
fig2 = plt.figure()
x = np.arange(-9, 10)
y = np.arange(-9, 10).reshape(-1, 1)
base = np.hypot(x, y)
ims = []
for add in np.arange(15):
ims.append((plt.pcolor(x, y, base + add, norm=plt.Normalize(0, 30)),))
im_ani = ArtistAnimation(fig2, ims, interval=50, repeat_delay=1000, blit=True)
# im_ani.save('im.mp4')
plt.show()
------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
