The basic functionality of the package is captured in the function `fit_dynamics`

which takes an input of type `Array{T,3} where T <: Real`

, which is assumed to have dimensions time by cells by trials. One can then do,

```
dims = size(X,2)
q = PopulationDynamics.fit_dynamics(X)
Aq = PopulationDynamics.sfunc(q.minimizer, dim)
```

which returns an `(dims+1)×(dims+1)`

transformation matrix `Aq`

which captures rotation, scaling and shearing in `Aq[1:dims-1, 1:dims-1]`

and translation in `Aq[1:end-1, end]`

.

08/08/2018

about 2 years ago

16 commits