automl (version 1.0.5)

autopar: parameters for automatic hyperparameters optimization

Description

List of parameters to allow multi deep neural network automatic hyperparameters tuning with Particle Swarm Optimization Not mandatory (the list is preset and all arguments are initialized with default value) but it is advisable to adjust some important arguments for performance reasons (including processing time)

Arguments

psopartpopsize

number of particles in swarm, the main argument that should be tuned #tuning priority 1

psoxxx

see pso for other PSO specific arguments details

numiterations

number of convergence steps between particles (hyperparameters), default value 3) #tuning priority 1

auto_modexec

if ‘TRUE’ the type of Neural Net optimization will be randomly choosen between ‘trainwgrad’ and ‘trainwpso’ for each particle default value is ‘FALSE’ (so default value of argument ‘modexec’ in automl_train_manual function) the value can be forced if defined in hpar list

auto_minibatchsize

see below

auto_minibatchsize_min

see below

auto_minibatchsize_max

‘auto_minibatch’ default value ‘TRUE’ for automatic adjustment of ‘minibatchsize’ argument in automl_train_manual function the minimum and maximum value for ‘minibatchsize’ corespond to 2 to the power value (default 0 for ‘auto_minibatchsize_min’ and 9 for ‘auto_minibatchsize_max’)

auto_learningrate

see below

auto_learningrate_min

see below

auto_learningrate_max

‘auto_learningrate’ default value ‘TRUE’ for automatic adjustment of ‘learningrate’ argument in automl_train_manual function the minimum and maximum value for ‘learningrate’ correspond to 10 to the power negative value (default -5 for ‘auto_learningrate_min’ and -2 for ‘auto_learningrate_max’)

auto_beta1

see below

auto_beta2

‘auto_beta1’ and ‘auto_beta2’ default value ‘TRUE’ for automatic adjustment of ‘beta1’ and ‘beta2’ argument in automl_train_manual function

auto_psopartpopsize

see below

auto_psopartpopsize_min

see below

auto_psopartpopsize_max

‘auto_psopartpopsize’ default value ‘TRUE’ for automatic adjustment of ‘psopartpopsize’ argument in automl_train_manual function the minimum and maximum value for ‘learningrate’ ; default 2 for ‘auto_psopartpopsize_min’ and 50 for ‘auto_psopartpopsize_max’)

seed

seed for reproductibility (default 4)

nbcores

number of cores used to parallelize particles optimization, not available on Windows (default 1, automatically reduced if not enough cores)

verbose

to display or not the costs at each iteration for each particle (default TRUE)