sktime.benchmarking.results

class sktime.benchmarking.results.HDDResults(path)[source]

Bases: sktime.benchmarking.base.HDDBaseResults

check_fitted_strategy_exists(strategy_name, dataset_name, cv_fold)[source]
check_predictions_exist(strategy_name, dataset_name, cv_fold, train_or_test)[source]
load_fitted_strategy(strategy_name, dataset_name, cv_fold)[source]

Load saved (fitted) strategy

load_predictions(cv_fold, train_or_test)[source]

Load saved predictions

save_fitted_strategy(strategy, dataset_name, cv_fold)[source]

Save fitted strategy

save_predictions(strategy_name, dataset_name, y_true, y_pred, y_proba, index, cv_fold, train_or_test, fit_estimator_start_time=None, fit_estimator_end_time=None, predict_estimator_start_time=None, predict_estimator_end_time=None)[source]

Saves the predictions of trained estimators.

Parameters
  • strategy_name (string) – Name of fitted strategy

  • dataset_name (string) – Name of dataset on which the strategy is fitted

  • y_true (numpy array) – array with true labels

  • y_pred (numpy array) – array of predicted labels

  • y_proba (numpy array) – array of probabilities associated with the predicted values

  • index (numpy array) – dataset indeces of the y_true data points

  • fit_estimator_start_time (pandas timestamp (default=None)) – timestamp when fitting the estimator began

  • fit_estimator_end_time (pandas timestamp (default=None)) – timestamp when fitting the estimator ended

  • predict_estimator_start_time (pandas timestamp (default=None)) – timestamp when the estimator began making predictions

  • predict_estimator_end_time (pandas timestamp (default=None)) – timestamp when the estimator finished making predictions

class sktime.benchmarking.results.RAMResults[source]

Bases: sktime.benchmarking.base.BaseResults

check_fitted_strategy_exists(strategy, dataset_name, cv_fold)[source]
check_predictions_exist(strategy, dataset_name, cv_fold, train_or_test)[source]
load_fitted_strategy(strategy_name, dataset_name, cv_fold)[source]

Load fitted strategies for all datasets and strategies iteratively

load_predictions(cv_fold, train_or_test)[source]

Loads predictions for all datasets and strategies iteratively

save()[source]

Save results object as master file

save_fitted_strategy(strategy, dataset_name, cv_fold)[source]
save_predictions(strategy_name, dataset_name, y_true, y_pred, y_proba, index, cv_fold, train_or_test, fit_estimator_start_time=None, fit_estimator_end_time=None, predict_estimator_start_time=None, predict_estimator_end_time=None)[source]

Saves the predictions of trained estimators.

Parameters
  • strategy_name (string) – Name of fitted strategy

  • dataset_name (string) – Name of dataset on which the strategy is fitted

  • y_true (numpy array) – array with true labels

  • y_pred (numpy array) – array of predicted labels

  • y_proba (numpy array) – array of probabilities associated with the predicted values

  • index (numpy array) – dataset indeces of the y_true data points

  • fit_estimator_start_time (pandas timestamp (default=None)) – timestamp when fitting the estimator began

  • fit_estimator_end_time (pandas timestamp (default=None)) – timestamp when fitting the estimator ended

  • predict_estimator_begin_time (pandas timestamp (default=None)) – timestamp when the estimator began making predictions

  • predict_estimator_end_time (pandas timestamp (default=None)) – timestamp when the estimator finished making predictions