TIM Matlab is one of the three ways to access the
functionality in TIM. It consists of a set of Matlab functions
together with a mex file implemented in C++.
The Matlab implementation consists mostly of redirection
functions that check the parameters at the Matlab’s side,
and forward the calls to the C++ implementation via the
tim_matlab mex file.
The C++ implementation mostly consists of interpreting the
data via correct data abstractions so that they can be
forwarded to the actual computation functions of TIM Core.
To obtain TIM Matlab, see the main page.
In the function names, the t suffix stands for a temporal version, while the p suffix stands for a partialized version (partial mutual information and partial transfer entropy). Because partialization and temporality are orthogonal concepts, there are also versions which have both. These are named with the pt suffix.
You can find function-specific documentation by using Matlab’s integrated help system. For example, for help on using the differential_entropy_kl function:
>> help differential_entropy_kl
DIFFERENTIAL_ENTROPY_KL
A differential entropy estimate from samples
using Kozachenko-Leonenko nearest neighbor estimator.
H = differential_entropy_kl(S, k)
where
...