Array2GIF: convert NumPy arrays to (animated) GIF
+-------------+--------------+--------------+
| |ising1| + |ising2| + |ising3| +
| T = 0.7 T_c + T = 0.88 T_c + T = 1.06 T_c +
+-------------+--------------+--------------+
Array2GIF provides a single top-level function, write_gif()
, to
write a 3-D NumPy array to a GIF, or a list of these arrays to an
animated GIF. It works for me - just small animations of thermodynamics
simulations - like the magnetization in the Ising model shown here.
Usage
Here is an example for a 2 pixel by 3 pixel animated GIF with
two frames, switching 5 frames per second. All animations from this
library will loop indefinitely.
.. code-block:: python
import numpy as np
from array2gif import write_gif
dataset = [
np.array([
[[255, 0, 0], [255, 0, 0]], # red intensities
[[0, 255, 0], [0, 255, 0]], # green intensities
[[0, 0, 255], [0, 0, 255]] # blue intensities
]),
np.array([
[[0, 0, 255], [0, 0, 255]],
[[0, 255, 0], [0, 255, 0]],
[[255, 0, 0], [255, 0, 0]]
])
]
write_gif(dataset, 'rgbbgr.gif', fps=5)
# or for just a still GIF
write_gif(dataset[0], 'rgb.gif')
Installation
Either: ::
pip install array2gif
or: ::
pip install git+https://github.com/tanyaschlusser/array2gif.git#egg=array2gif
.. _the repository
: http://github.com/tanyaschlusser/array2gif
.. |ising1| image:: https://tanyaschlusser.github.io/ising/img/ising_animation_1.6.gif
:scale: 200%
:width: 84px
:align: middle
:alt: Animation of random pixels converging from two colors to one color. T = 0.7 T_c.
.. |ising2| image:: https://tanyaschlusser.github.io/ising/img/ising_animation_2.0.gif
:scale: 200%
:width: 84px
:align: middle
:alt: Animation of random pixels converging less slowly to one color. T = 0.88 T_c.
.. |ising3| image:: https://tanyaschlusser.github.io/ising/img/ising_animation_2.4.gif
:scale: 200%
:width: 84px
:align: middle
:alt: Animation of random pixels staying mostly random. T = 1.06 T_c.
.. :changelog:
Release History
1.0.4 (2018-06-22)
++++++++++++++++++
Bugfixes
-
Fixed issue where error was raised when non-integer dataset
is used: instead now it is cast to uint8
and a warning is
raised if there's a difference between the two images.
-
Addressed issue where no error is raised when there are
more colors than possible in a GIF (256 max). Although
it is possible to have a separate palette for every
single frame, this is not implemented right now, and
the initial global palette is used for the entire animation.
1.0.3 (2018-06-02)
++++++++++++++++++
Bugfixes
- Did not bump version in documentation last times.
1.0.1 (2018-06-02)
++++++++++++++++++
Bugfixes
- Fixed issue where the width and height of an image are swapped.
This is clear in image editors but was not obvious when viewed
in Chrome.
1.0.0 (2018-05-23)
++++++++++++++++++
Improvements
- It is now possible to use PIL ordering of data (rows x cols x rgb)
in addition to the original ordering (rgb x rows x cols).
Bugfixes
- Fixed issue where array2gif would raise
ValueError
when using
a 4D NumPy array with perfectly valid data. Now it is possible to
use either a list of 3D NumPy arrays, or a 4D NumPy array for
animated gifs.
0.1.0 (2016-09-30)
++++++++++++++++++
Initial release
- One single function,
write_gif
.