tigerforecast.experiments.load_prob_method_to_result¶
-
tigerforecast.experiments.
load_prob_method_to_result
(problem_ids=['ARMA-v0', 'Crypto-v0', 'SP500-v0'], method_ids=['LastValue', 'AutoRegressor', 'RNN', 'LSTM'], problem_to_methods=None, metrics='mse')[source]¶ Description: Initializes the experiment instance.
Parameters: - problem_ids (list) – ids of problems to evaluate on
- method_ids (list) – ids of methods to use
- problem_to_methods (dict) – map of the form problem_id -> list of method_id. If None, then we assume that the user wants to test every method in method_to_params against every problem in problem_to_params
- metrics – metrics to load