ForeTiS.utils.helper_functions
Module Contents
Functions
Create a list of all implemented models based on files existing in 'model' subdirectory of the repository. |
|
Get a mapping from model name (naming in package model without .py) to class name. |
|
|
Set all seeds of libs with a specific function for reproducibility of results |
|
Get the indexes for cv |
|
Get all datasplit info for a result folder |
- ForeTiS.utils.helper_functions.get_list_of_implemented_models()
Create a list of all implemented models based on files existing in ‘model’ subdirectory of the repository.
- Return type:
- ForeTiS.utils.helper_functions.get_mapping_name_to_class()
Get a mapping from model name (naming in package model without .py) to class name.
- Returns:
dictionary with mapping model name to class name
- Return type:
- ForeTiS.utils.helper_functions.set_all_seeds(seed=42)
Set all seeds of libs with a specific function for reproducibility of results
- Parameters:
seed (int) – seed to use
- ForeTiS.utils.helper_functions.get_indexes(df, datasplit, folds, seasonal_valtest, valtest_seasons, seasonal_periods, val_set_size_percentage)
Get the indexes for cv
- Parameters:
df (pandas.DataFrame) – data that should be splited
datasplit (str) – splitting method
folds (int) – number of folds of the hyperparameter optimization
valtest_seasons (int) – the number of seasons to be used for validation and testing when seasonal_valtest is True
seasonal_periods (int) – how many datapoints one season has
test_set_size_percentage – size of the test set relevant for cv-test and train-val-test
val_set_size_percentage (int) – size of the validation set relevant for train-val-test
- Returns:
train and test indexes
- ForeTiS.utils.helper_functions.get_datasplit_config_info_for_resultfolder(resultfolder)
Get all datasplit info for a result folder