Returns the calculated width parameter, but does not change with current parameter.
The training takes a few tens of minutes with current parameter settings.
New IL groups were added into the current parameter matrix.
Computational efficiency: Straight forward matrix methods rely on the eigendecomposition of the current parameter matrix during the parameter update.
The current parameter values for the models are included, as well as information on the most recent fit.
Here, a formulation is proposed that addresses these limitations of current parameter estimation methods.
Metropolis jumps to the new set of parameters from the current parameter set, while Metropolis-Hastings jumps always from the best fit.
For each model, the current parameter values are returned (including minimum and maximum values), along with parameter type (thawed or frozen) and units.
County names, size, and current parameter values appear in the black box above the timeline when the mouse scrolls over a county boundary.
The current parameter optimization methods either are easy to fall into local optimal solution, or are time consuming.
It would be interesting to plot performance as a function of K. Doing so will require a fair amount of CPU time, maybe a few days or so with my current parameter values.
