def test_scenario5(self): """ Scenario: Successfully building multi-label evaluations from models retrieved by tag Given I have previously executed "<scenario>" or reproduce it with arguments <kwargs> And I create BigML multi-label resources using models tagged as "<tag>" to evaluate and log evaluation in "<output>" And I check that the <number_of_labels> evaluations have been created And I check that the evaluation is ready Then the evaluation key "<key>" value for the model is greater than <value> Examples: |scenario | kwargs | tag | number_of_labels | output |key | value | scenario_ml_e1| {"tag": "my_multilabel_e_1", "data": "../data/multilabel.csv", "label_separator": ":", "number_of_labels": 7, "training_separator": ",", "output": "./scenario_ml_e1/evaluation"} | my_multilabel_e_1 | 7 | ./scenario_ml_e5/evaluation | average_phi | 0.8180 """ print self.test_scenario5.__doc__ examples = [[ 'scenario_ml_e1', '{"tag": "my_multilabel_e_1", "data": "data/multilabel.csv", "label_separator": ":", "number_of_labels": 7, "training_separator": ",", "output": "scenario_ml_e1/evaluation"}', 'my_multilabel_e_1', '7', 'scenario_ml_e5/evaluation', 'average_phi', '0.8180' ]] for example in examples: print "\nTesting with:\n", example test_pred.i_have_previous_scenario_or_reproduce_it( self, example[0], example[1]) ml_eval.i_create_ml_evaluations_from_tagged_models( self, tag=example[2], output=example[4]) test_pred.i_check_create_evaluations( self, number_of_evaluations=example[3]) ml_eval.i_check_evaluation_ready(self) evaluation.i_check_evaluation_key(self, key=example[5], value=example[6])
def test_scenario7(self): """ Scenario: Successfully building ensemble evaluations from start and test-split: Given I create BigML resources uploading train "<data>" file to evaluate an ensemble of <number_of_models> models with test-split <split> and log evaluation in "<output>" And I check that the source has been created And I check that the dataset has been created And I check that the train dataset has been created And I check that the test dataset has been created And I check that the ensemble has been created And I check that the evaluation has been created Then the evaluation key "<key>" value for the model is greater than <value> And I evaluate the ensemble in directory "<directory>" with the dataset in directory "<directory>" and log evaluation in "<output2>" And I check that the evaluation has been created Then the evaluation key "<key>" value for the model is greater than <value> Examples: | data | output | split | number_of_models | key | value | directory | output2 | ../data/iris.csv | ./scenario_e8/evaluation | 0.2 | 5 | average_phi | 0.94 | ./scenario_e8/ | ./scenario_e9/evaluation """ print self.test_scenario7.__doc__ examples = [ ['data/iris.csv', 'scenario_e8/evaluation', '0.2', '5', 'average_phi', '0.94', 'scenario_e8', 'scenario_e9/evaluation']] for example in examples: print "\nTesting with:\n", example evaluation.i_create_with_split_to_evaluate_ensemble(self, data=example[0], number_of_models=example[3], split=example[2], output=example[1]) test_pred.i_check_create_source(self) test_pred.i_check_create_dataset(self, suffix=None) test_pred.i_check_create_dataset(self, suffix='train ') test_pred.i_check_create_dataset(self, suffix='test ') test_pred.i_check_create_ensemble(self) test_pred.i_check_create_evaluation(self) evaluation.i_check_evaluation_key(self, key=example[4], value=example[5]) evaluation.i_evaluate_ensemble_with_dataset(self, ensemble_dir=example[6], dataset_dir=example[6], output=example[7]) test_pred.i_check_create_evaluation(self) evaluation.i_check_evaluation_key(self, key=example[4], value=example[5])
def test_scenario6(self): """ Scenario: Successfully building evaluations from start and test-split: Given I create BigML resources uploading train "<data>" file to evaluate with test-split <split> and log evaluation in "<output>" And I check that the source has been created And I check that the dataset has been created And I check that the train dataset has been created And I check that the test dataset has been created And I check that the model has been created And I check that the evaluation has been created Then the evaluation key "<key>" value for the model is greater than <value> Examples: | data | output | split | key | value | | ../data/iris.csv | ./scenario_e6/evaluation | 0.2 | average_phi | 0.85 | """ print self.test_scenario6.__doc__ examples = [ ['data/iris.csv', 'scenario_e6/evaluation', '0.2', 'average_phi', '0.85']] for example in examples: print "\nTesting with:\n", example evaluation.i_create_with_split_to_evaluate(self, data=example[0], split=example[2], output=example[1]) test_pred.i_check_create_source(self) test_pred.i_check_create_dataset(self, suffix=None) test_pred.i_check_create_dataset(self, suffix='train ') test_pred.i_check_create_dataset(self, suffix='test ') test_pred.i_check_create_model(self) test_pred.i_check_create_evaluation(self) evaluation.i_check_evaluation_key(self, key=example[3], value=example[4])
def test_scenario7(self): """ Scenario: Successfully building ensemble evaluations from start and test-split: Given I create BigML resources uploading train "<data>" file to evaluate an ensemble of <number_of_models> models with test-split <split> and log evaluation in "<output>" And I check that the source has been created And I check that the dataset has been created And I check that the train dataset has been created And I check that the test dataset has been created And I check that the ensemble has been created And I check that the evaluation has been created Then the evaluation key "<key>" value for the model is greater than <value> And I evaluate the ensemble in directory "<directory>" with the dataset in directory "<directory>" and log evaluation in "<output2>" And I check that the evaluation has been created Then the evaluation key "<key>" value for the model is greater than <value> Examples: | data | output | split | number_of_models | key | value | directory | output2 | ../data/iris.csv | ./scenario_e8/evaluation | 0.2 | 5 | average_phi | 0.94 | ./scenario_e8/ | ./scenario_e9/evaluation """ print self.test_scenario7.__doc__ examples = [[ 'data/iris.csv', 'scenario_e8/evaluation', '0.2', '5', 'average_phi', '0.94', 'scenario_e8', 'scenario_e9/evaluation' ]] for example in examples: print "\nTesting with:\n", example evaluation.i_create_with_split_to_evaluate_ensemble( self, data=example[0], number_of_models=example[3], split=example[2], output=example[1]) test_pred.i_check_create_source(self) test_pred.i_check_create_dataset(self, suffix=None) test_pred.i_check_create_dataset(self, suffix='train ') test_pred.i_check_create_dataset(self, suffix='test ') test_pred.i_check_create_ensemble(self) test_pred.i_check_create_evaluation(self) evaluation.i_check_evaluation_key(self, key=example[4], value=example[5]) evaluation.i_evaluate_ensemble_with_dataset( self, ensemble_dir=example[6], dataset_dir=example[6], output=example[7]) test_pred.i_check_create_evaluation(self) evaluation.i_check_evaluation_key(self, key=example[4], value=example[5])
def test_scenario4(self): """ Scenario: Successfully building multi-label evaluations from models file Given I have previously executed "<scenario>" or reproduce it with arguments <kwargs> And I create BigML multi-label resources using models in file "<models_file>" to evaluate and log evaluation in "<output>" And I check that the <number_of_labels> evaluations have been created And I check that the evaluation is ready Then the evaluation key "<key>" value for the model is greater than <value> Examples: |scenario | kwargs | models_file | number_of_labels | output |key | value | scenario_ml_e1| {"tag": "my_multilabel_e_1", "data": "../data/multilabel.csv", "label_separator": ":", "number_of_labels": 7, "training_separator": ",", "output": "./scenario_ml_e1/evaluation"} | ./scenario_ml_e1/models | 7 | ./scenario_ml_e4/evaluation | average_phi | 0.8180 """ print self.test_scenario4.__doc__ examples = [ ['scenario_ml_e1', '{"tag": "my_multilabel_e_1", "data": "data/multilabel.csv", "label_separator": ":", "number_of_labels": 7, "training_separator": ",", "output": "scenario_ml_e1/evaluation"}', 'scenario_ml_e1/models', '7', 'scenario_ml_e4/evaluation', 'average_phi', '0.8180']] for example in examples: print "\nTesting with:\n", example test_pred.i_have_previous_scenario_or_reproduce_it(self, example[0], example[1]) ml_eval.i_create_ml_evaluations_from_models(self, models_file=example[2], output=example[4]) test_pred.i_check_create_evaluations(self, number_of_evaluations=example[3]) ml_eval.i_check_evaluation_ready(self) evaluation.i_check_evaluation_key(self, key=example[5], value=example[6])
def test_scenario6(self): """ Scenario: Successfully building evaluations from start and test-split: Given I create BigML resources uploading train "<data>" file to evaluate with test-split <split> and log evaluation in "<output>" And I check that the source has been created And I check that the dataset has been created And I check that the train dataset has been created And I check that the test dataset has been created And I check that the model has been created And I check that the evaluation has been created Then the evaluation key "<key>" value for the model is greater than <value> Examples: | data | output | split | key | value | | ../data/iris.csv | ./scenario_e6/evaluation | 0.2 | average_phi | 0.85 | """ print self.test_scenario6.__doc__ examples = [[ 'data/iris.csv', 'scenario_e6/evaluation', '0.2', 'average_phi', '0.85' ]] for example in examples: print "\nTesting with:\n", example evaluation.i_create_with_split_to_evaluate(self, data=example[0], split=example[2], output=example[1]) test_pred.i_check_create_source(self) test_pred.i_check_create_dataset(self, suffix=None) test_pred.i_check_create_dataset(self, suffix='train ') test_pred.i_check_create_dataset(self, suffix='test ') test_pred.i_check_create_model(self) test_pred.i_check_create_evaluation(self) evaluation.i_check_evaluation_key(self, key=example[3], value=example[4])