OptimResult

class liesel.goose.optim.OptimResult(model_state, position, iteration, iteration_best, history, max_iter, n_train, n_validation)

Bases: object

Holds the results of model optimization with optim_flat().

Attributes

This section is empty if this class has only inherited attributes.

model_state	Final model state after optimization.
position	Position dictionary of optimized parameters with their final values.
iteration	Iteration counter of the last iteration.
iteration_best	Iteration counter of the iteration with lowest loss.
history	History of loss evaluations and, if applicable, intermediate position values.
max_iter	Maximum number of iterations.
n_train	Number of training observations.
n_validation	Number of test observations.