add_experiment_if_not_exists: Add experiment if not exists

Description Usage Arguments Value

Description

Check if experiment exists, verifies parameters, creates data to create_experiment function and finally starts creation of MLJAR experiment.

Usage

1
2
3
4
add_experiment_if_not_exists(project_hid, train_dataset, valid_dataset,
  experiment_title, project_task, validation_kfolds, validation_shuffle,
  validation_stratify, validation_train_split, algorithms, metric, tuning_mode,
  time_constraint, create_ensemble)

Arguments

project_hid

character with project identifier

train_dataset

character with path to training dataset

valid_dataset

character with path to validation dataset

experiment_title

character with experiment title

project_task

character with project task

validation_kfolds

number of folds to be used in validation

validation_shuffle

boolean which specify if shuffle samples before training

validation_stratify

boolean which decides whether samples will be divided into folds with the same class distribution

validation_train_split

ratio how to split training dataset into train and validation

algorithms

list of algorithms to use

metric

charcater with metric

tuning_mode

tuning mode

time_constraint

numeric with time limit to calculate algorithm

create_ensemble

whether or not to create ensemble

Value

experiment details structure


mljar documentation built on May 2, 2019, 2:50 p.m.