KaneAI - World’s First E2E Software Testing Agent.Book a Demo Join TestMu Conference

How to use rule_wrapper method in hypothesis

Best Python code snippet using hypothesis

run_exp_stub.py

Source:run_exp_stub.py

1import os2import random3from typing import List, Optional4import pandas as pd5from pylo.language.lp import (Clause as PyloClause, global_context as pylo_global_context)6from tqdm import tqdm7from artificial_bias_experiments.experiment_utils import print_or_log8from kbc_pul.amie.amie_output_rule_extraction import get_amie_rules_from_rule_tsv_file9from artificial_bias_experiments.evaluation.ground_truth_utils import TrueEntitySetsTuple, \10    get_true_entity_sets_as_string_sets11from artificial_bias_experiments.noisy_prop_scores.available_prop_scores_to_pu_metrics_controller import \12    RuleWrapperNoisyPropScoresToPuMetricsMap13from artificial_bias_experiments.noisy_prop_scores.scar.experiment_info import \14    NoisyPropScoresSCARExperimentInfo15from artificial_bias_experiments.noisy_prop_scores.scar.noisy_prop_scores_scar_file_naming import \16    NoisyPropScoresSCARFileNamer17from kbc_pul.data_structures.pandas_kb import PandasKnowledgeBaseWrapper18from kbc_pul.data_structures.rule_wrapper import RuleWrapper, filter_rules_predicting19from kbc_pul.experiments_utils.load_df_ground_truth import get_df_ground_truth20from kbc_pul.observed_data_generation.abstract_triple_selection import ObservedTargetRelationInfo21from kbc_pul.observed_data_generation.sar_two_subject_groups.sar_two_subject_groups_prop_scores import \22    PropScoresTwoSARGroups23from kbc_pul.observed_data_generation.scar.scar_propensity_score_controller import \24    SCARPropensityScoreController25from kbc_pul.observed_data_generation.scar.scar_triple_selection import SCARTripleSelector26from kbc_pul.rule_metrics.prediction_cache_rule_metrics.rule_cwa_and_pca_confidences_from_cached_predictions import \27    set_rule_wrapper_cwa_and_pca_confidence_calculated_from_cache28from kbc_pul.rule_metrics.prediction_cache_rule_metrics.rule_ipw_and_ipw_pca_confidences_from_cached_predictions import \29    calculate_rule_ipw_and_ipw_pca_confidences_from_df_cached_predictions30from kbc_pul.rule_metrics.prediction_cache_rule_metrics.rule_true_confidence_on_observed_data_from_cached_predictions import \31    calculate_true_confidence_metrics_from_df_cached_predictions32def write_observed_target_relation_to_csv(33        mask_observed_rows: pd.Series,34        experiment_dir: str,35        experiment_info: NoisyPropScoresSCARExperimentInfo,36        random_trial_index: int37):38    filename_mask_observed_rows: str = os.path.join(39        experiment_dir,40        f"mask_{experiment_info.target_relation}"41        f"c{experiment_info.true_label_frequency}_trial{random_trial_index}.csv"42    )43    mask_observed_rows.astype(int).to_csv(44        filename_mask_observed_rows,45        sep="\t",46        index=False,47        header=False48    )49def run_single_experiment_setting_of_experiment_noisy_prop_scores_scar(50        filename_ground_truth_dataset: str,51        separator_ground_truth_dataset: str,52        amie_rule_tsv_filename: str,53        experiment_info: NoisyPropScoresSCARExperimentInfo,54        random_seed: int,55        n_random_trials: int,56        verbose: bool = False,57):58    experiment_dir: str = NoisyPropScoresSCARFileNamer.get_dir_experiment_specific(59        experiment_info=experiment_info60    )61    print(experiment_dir)62    if verbose:63        print(experiment_dir)64    rng = random.Random(random_seed)65    df_ground_truth: pd.DataFrame = get_df_ground_truth(66        filename_ground_truth_dataset, separator_ground_truth_dataset67    )68    print_or_log(message="Generating SCAR dataset", logger=None, verbose=verbose)69    pandas_kb_wrapper = PandasKnowledgeBaseWrapper.create_from_full_data(df_full_data=df_ground_truth)70    df_ground_truth_target_relation: pd.DataFrame = pandas_kb_wrapper.get_relation(71        experiment_info.target_relation72    )73    if verbose:74        print(f"ground truth:")75        print(f"\t{df_ground_truth.shape[0]} literals")76        print(f"\t{df_ground_truth_target_relation.shape[0]} {experiment_info.target_relation} (target) literals")77    dir_rule_wrappers: str = os.path.join(78        experiment_dir,79        'rule_wrappers'80    )81    if not os.path.exists(dir_rule_wrappers):82        os.makedirs(dir_rule_wrappers)83    triple_selector: SCARTripleSelector = SCARTripleSelector(84        constant_label_frequency=experiment_info.true_label_frequency,85        verbose=False86    )87    #############################################################################################88    random_trial_index: int89    for random_trial_index in range(n_random_trials):90        observed_target_relation_info: ObservedTargetRelationInfo = triple_selector.select_observed_target_relation(91            df_ground_truth_target_relation=df_ground_truth_target_relation,92            rng=rng93        )94        write_observed_target_relation_to_csv(95            mask_observed_rows=observed_target_relation_info.mask,96            experiment_dir=experiment_dir,97            experiment_info=experiment_info,98            random_trial_index=random_trial_index99        )100        # Set the observed target relation101        pandas_kb_wrapper.replace_predicate(102            relation=experiment_info.target_relation,103            new_df_for_relation=observed_target_relation_info.df104        )105        print_or_log(message="Finished generating biased dataset (non-PCA)", logger=None, verbose=verbose)106        print_or_log(message="Evaluating rules on biased dataset (non-PCA)", logger=None, verbose=verbose)107        true_ent_sets_tuple: TrueEntitySetsTuple = get_true_entity_sets_as_string_sets(108            df_ground_truth_target_relation109        )110        #############################################################################################111        if verbose:112            print("Start evaluation phase")113            print(f"True Label frequency: {experiment_info.true_label_frequency}")114        rule_wrapper_list: List[RuleWrapper] = [115            rule_wrapper.clone_with_metrics_unset()116            for rule_wrapper in filter_rules_predicting(117                get_amie_rules_from_rule_tsv_file(amie_rule_tsv_filename),118                head_functor_set={119                    experiment_info.target_relation120                }121            )122        ]123        print(f"Start evaluation (random trial index {random_trial_index}"124              f"  ({random_trial_index + 1} / {n_random_trials}))")125        rule_wrapper: RuleWrapper126        for rule_wrapper in tqdm(rule_wrapper_list, disable=not verbose):127            rule: PyloClause = rule_wrapper.rule128            if verbose:129                print(f"Rule: {rule}")130            o_df_cached_predictions: Optional[pd.DataFrame] = pandas_kb_wrapper.calculate_prediction_cache_for_rule(131                rule=rule_wrapper.rule132            )133            if o_df_cached_predictions is None or (len(o_df_cached_predictions) == 0):134                if verbose:135                    print(f"ZERO PREDICTIONS for rule {rule}")136            else:137                # STD, PCA138                set_rule_wrapper_cwa_and_pca_confidence_calculated_from_cache(rule_wrapper, o_df_cached_predictions)139                # TRUE CONF & CONF*140                calculate_true_confidence_metrics_from_df_cached_predictions(141                    rule_wrapper=rule_wrapper,142                    df_cached_predictions=o_df_cached_predictions,143                    df_ground_truth_target_relation=df_ground_truth_target_relation,144                    true_entity_sets=true_ent_sets_tuple145                )146                filename_rule_wrapper: str = os.path.join(147                    dir_rule_wrappers,148                    f"{str(rule_wrapper.rule)}_trial{random_trial_index}.json.gz"149                )150                rule_wrapper.to_json_file(filename_rule_wrapper)151                ##################################################152                dir_pu_metrics_of_rule_wrappers: str = os.path.join(153                    dir_rule_wrappers,154                    "pu_metrics"155                )156                if not os.path.exists(dir_pu_metrics_of_rule_wrappers):157                    os.makedirs(dir_pu_metrics_of_rule_wrappers)158                filename_pu_metrics_of_rule_wrapper: str = os.path.join(159                    dir_pu_metrics_of_rule_wrappers,160                    f"{str(rule_wrapper.rule)}_trial{random_trial_index}.tsv.gz"161                )162                noisy_prop_score_tuple_to_pu_metrics_map_controller = RuleWrapperNoisyPropScoresToPuMetricsMap(163                    rule_str=str(rule_wrapper.rule),164                    random_trial_index=random_trial_index,165                    true_prop_scores=PropScoresTwoSARGroups(166                        in_filter=experiment_info.true_label_frequency,167                        other=experiment_info.true_label_frequency168                    )169                )170                noisy_propensity_score_other_entities: float171                for noisy_label_frequency in experiment_info.available_label_frequency_list:172                    rule_wrapper.o_relative_pu_confidence_unbiased = None173                    rule_wrapper.o_relative_pu_confidence_pca_subject_to_object = None174                    rule_wrapper.o_relative_pu_confidence_pca_object_to_subject = None175                    rule_wrapper.o_c_weighted_std_conf = None176                    available_label_frequency_scar_propensity_score_controller = SCARPropensityScoreController(177                        constant_label_frequency=noisy_label_frequency178                    )179                    calculate_rule_ipw_and_ipw_pca_confidences_from_df_cached_predictions(180                        rule_wrapper=rule_wrapper,181                        df_cached_predictions=o_df_cached_predictions,182                        pylo_context=pylo_global_context,183                        propensity_score_controller=available_label_frequency_scar_propensity_score_controller,184                        verbose=verbose185                    )186                    rule_wrapper.set_inverse_c_weighted_std_confidence(187                        label_frequency=noisy_label_frequency188                    )189                    noisy_prop_score_tuple_to_pu_metrics_map_controller.add_pu_metrics_for_available_prop_scores(190                        noisy_prop_scores=PropScoresTwoSARGroups(191                            in_filter=noisy_label_frequency,192                            other=noisy_label_frequency193                        ),194                        o_relative_pu_confidence_unbiased=rule_wrapper.o_relative_pu_confidence_unbiased,195                        o_relative_pu_confidence_pca_subject_to_object=rule_wrapper.o_relative_pu_confidence_pca_subject_to_object,196                        o_relative_pu_confidence_pca_object_to_subject=rule_wrapper.o_relative_pu_confidence_pca_object_to_subject,197                        o_inverse_c_weighted_std_confidence=rule_wrapper.o_c_weighted_std_conf198                    )199                noisy_prop_score_tuple_to_pu_metrics_map_controller.to_tsv(200                    filename_noisy_prop_scores_to_pu_metrics_map=filename_pu_metrics_of_rule_wrapper201                )...

rule.py

Source:rule.py

1from flask import jsonify, request2from wrappers import rule_wrapper3from validators import rule_validator4from utils.nft_errors import abort, NFTError, NFTValidationError5def rules():6    '''7    GET:8      List all rules in the system9    POST:10      Create a new rule in the system11    '''12    if request.method == 'POST':13        try:14            rule_json = rule_validator.validate_new_rule(request.get_json())15            rule = rule_wrapper.create_rule(rule_json)16            response = jsonify(rule=rule)17            response.status_code = 20118            return response19        except NFTValidationError as e:20            return abort(400, e)21        except NFTError as e:22            return abort(500, e)23    else:24        return jsonify(rules=rule_wrapper.list_all_rules())25def rule(rule_id):26    '''27    GET:28      Get a rule by it's id29    DELETE:30      Delete the rule with the specified id31    '''32    if request.method == 'DELETE':33        try:34            rule = rule_validator.validate_rule_delete(rule_id)35            rule = rule_wrapper.delete_rule(rule_id)36            response = jsonify({})37            response.status_code = 20438            return response39        except NFTValidationError as e:40            return abort(400, e)41        except NFTError as e:42            return abort(500, e)43    else:...

rule_validator.py

Source:rule_validator.py

1from utils.nft_errors import NFTValidationError, abort2from wrappers import rule_wrapper3def validate_new_rule(rule_json):4    rule = rule_json.pop('rule', None)5    validation_error = NFTValidationError('rule')6    # JSON errors7    if not rule:8        validation_error.add_error('rule', 'No "rule" field in rule json')9        raise validation_error10    # Return11    if validation_error.has_errors():12        raise validation_error13    else:14        return rule15def validate_rule_delete(rule_id):16    if not rule_id:17        pass18    rule = rule_wrapper.get_rule(rule_id)19    if not rule:20        pass...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.