number of particles in swarm (discrete value)
(&#8216;autopar&#8217; context: default value 8, which means that 8 different neural net hyperparameters sets will be tested
(&#8216;hpar&#8217; context: default value 50, which means that 50 neural net weights sets will be tested
#tuning priority 1

psopartpopsize

Minimum value for particles positions (default value -10)

psovarvalmin

maximum value for particles positions (default value 10)

psovarvalmax

ratio applied to limit velocities (continuous value between 0 and 1, default value 0.2)

psovelocitymaxratio

inertia damp ratio (continuous value between 0 and 1, default value 1 equivalent to OFF)

psoinertiadampratio

psokappa

psophi1

psophi2

List of parameters and hyperparameters for Particle Swarm Optimization

Fits from simple regression to highly customizable deep neural networks
either with gradient descent or metaheuristic, using automatic hyper parameters
tuning and custom cost function.
A mix inspired by the common tricks on Deep Learning and Particle Swarm Optimization.

pso: PSO parameters and hyperparameters

Description

Arguments

See Also