Information Profiles¶
There are several ways to decompose the information contained in a joint distribution. Here, we will demonstrate their behavior using four examples drawn from [ASBY14]:
In [1]: In [1]: from dit.profiles import *
Shannon Partition and Extropy Partition¶
The I-diagrams, or ShannonPartition
, for these four examples can be computed thusly:
In [2]: In [6]: ShannonPartition(ex1)
---------------------------------------------------------------------------
NameError Traceback (most recent call last)
<ipython-input-2-b4c036d85de7> in <module>
----> 1 ShannonPartition(ex1)
NameError: name 'ex1' is not defined
In [3]: +----------+--------+
...: | measure | bits |
...: +----------+--------+
...: | H[0|1,2] | 0.103 |
...: | H[1|0,2] | 0.103 |
...: | H[2|0,1] | 0.103 |
...: | I[0:1|2] | 0.142 |
...: | I[0:2|1] | 0.142 |
...: | I[1:2|0] | 0.142 |
...: | I[0:1:2] | 0.613 |
...: +----------+--------+
...: