rose.optimisation package
Submodules
rose.optimisation.optimisation module
- class rose.optimisation.optimisation.ModelResults
Bases:
object
- class rose.optimisation.optimisation.Optimisation
Bases:
object- initialise()
Initialise optimisation :return:
- least_square(x0, method='lm', ftol=1e-08, xtol=1e-08, gtol=1e-08)
Solve a nonlinear least-squares problem
- Parameters:
x0 – initial parameters
method – method of optimisation method, “trf”, “dogbox”, “lm” trf is best for large bounded sparse problems lm is best for smaller unbounded non sparsed matrices
ftol – tolerance of cost function return
xtol – tolerance of variable change
gtol – tolerance of norm of the gradient
- Returns:
- reset_model()
Reset numerical model :return:
- residual_function(parameters, method='maximum')
computes the minimisation function
- Parameters:
parameters – ordered 1d-array of input parameters
- Returns:
- class rose.optimisation.optimisation.OptimisationModelPart
Bases:
object
rose.optimisation.particle_swarm_optimisation module
- class rose.optimisation.particle_swarm_optimisation.PSO(costFunc, x0, bounds, num_particles, maxiter)
Bases:
object
- class rose.optimisation.particle_swarm_optimisation.Particle(x0)
Bases:
object- evaluate(costFunc)
- update_position(bounds)
- update_velocity(pos_best_g)
- rose.optimisation.particle_swarm_optimisation.func1(x)