autopar.Rd
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)
psopartpopsize | number of particles in swarm, the main argument that should be tuned (default value 8, which is quite low) |
---|---|
psoxxx | see pso for other PSO specific arguments details |
numiterations | number of convergence steps between particles (hyperparameters), default value 3) |
auto_modexec | if ‘TRUE’ the type of Neural Net optimization will be randomly choosen between ‘trainwgrad’ and ‘trainwpso’ for each particle |
auto_runtype | if ‘2steps’ the 2 following steps will be run automatically (default value is ‘normal’): |
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 |
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 |
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 (concern only ‘modexec’ set to ‘trainwpso’) |
auto_lambda | see below |
auto_lambda_min | see below |
auto_lambda_max | ‘auto_lambda’ default value ‘FALSE’ for automatic adjustment of ‘lambda’ regularization argument in automl_train_manual function |
auto_psovelocitymaxratio | see below |
auto_psovelocitymaxratio_min | see below |
auto_psovelocitymaxratio_max | ‘auto_psovelocitymaxratio’ default value ‘TRUE’ for automatic adjustment of ‘psovelocitymaxratio’ PSO velocity max ratio argument in automl_train_manual function |
auto_layers | see below (‘auto_layers’ default value ‘TRUE’ for automatic adjustment of layers shape in automl_train_manual function) |
auto_layers_min | (linked to ‘auto_layers’ above, set hpar ‘layersshape’ and ‘layersacttype’) the minimum number of hidden layers (default 1 no hidden layer) |
auto_layers_max | (linked to ‘auto_layers’ above, set hpar ‘layersshape’ and ‘layersacttype’) the maximum number of hidden layers (default 2) |
auto_layersnodes_min | (linked to ‘auto_layers’ above, set hpar ‘layersshape’ and ‘layersacttype’) the minimum number of nodes per layer (default 3) |
auto_layersnodes_max | (linked to ‘auto_layers’ above, set hpar ‘layersshape’ and ‘layersacttype’) the maximum number of nodes per layer (default 33) |
auto_layersdropo | see below |
auto_layersdropoprob_min | see below |
auto_layersdropoprob_max | ‘auto_layersdropo’ default value ‘FALSE’ for automatic adjustment of hpar ‘layersdropoprob’ in automl_train_manual function) |
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) |
subtimelimit | time limit in seconds for sub modelizations to avoid waiting too long for a specific particle to finish its modelization (default 3600) |