Defaults

In this section, we explain which parameters are set by MLAir during the ExperimentSetup if not specified by the user. This is important information for example if a new Custom Data Handler is implemented.

Defaults Overview

parameter	default	comment
batch_path
batch_size	512
bootstrap_path
competitor_path
competitors	[]
create_new_bootstraps	False
create_new_model	True
data_handler	DefaultDataHandler
data_origin
data_path
debug	-	MLAir checks if it is running in debug mode and stores this dimensions
end	“2017-12-31”
epochs	20	This is just a placeholder to prevent unintended longish training
evaluate_bootstraps	True	Bootstrapping may take some time.
experiment_name
experiment_path
extreme_values	None
extremes_on_right_tail_only	False	Could be used for skewed distributions
forecast_path
fraction_of_training
hostname
hpc_hosts
interpolation_limit	1
interpolation_method	“linear”
logging_path
login_nodes
model_class
model_path
neighbors
number_of_bootstraps
overwrite_local_data
permute_data	False
plot_list
plot_path
start	“1997-01-01”
stations
statistics_per_var
target_dim
target_var
test_start
test_end
test_min_length
time_dim
train_model
train_end
train_min_length
train_start
transformation	{}	implement all further transformation functionality inside your custom data handler
use_all_stations_on_all_data_sets
use_multiprocessing	True	is used if MLAir is not running in debug mode
use_multiprocessing_on_debug	False	is used if MLAir is running in debug mode
upsampling
val_end
val_min_length
val_start
variables
window_history_size	13
window_lead_time	3