main.py

import sys

sys.path.append('./src')
from dgld.utils.evaluation import split_auc, score_other, curve_obtain, split_ap
from dgld.utils.common import seed_everything
from dgld.utils.argparser import parse_all_args
from dgld.utils.load_data import load_data, load_custom_data, load_truth_data
from dgld.utils.inject_anomalies import inject_contextual_anomalies, inject_structural_anomalies
from dgld.utils.common_params import Q_MAP, K, P
from dgld.utils.log import Dgldlog
from dgld.models import *
import numpy as np
import os

truth_list = ['weibo', 'reddit2']

if __name__ == "__main__":
    args_dict, args = parse_all_args()

    print(args_dict, flush=True)

    data_name = args_dict['dataset']
    save_path = args.save_path
    exp_name = args.exp_name
    log = Dgldlog(save_path, exp_name, args)

    res_list_final = []
    res_list_attrb = []
    res_list_struct = []
    res_list_ap = []
    res_list_attrb_ap = []
    res_list_struct_ap = []

    seed_list = [i for i in range(1, args.runs + 1)]
    for runs in range(args.runs):
        log.update_runs()
        seed = seed_list[runs]
        seed_everything(seed)
        args_dict['seed'] = seed

        if data_name in truth_list:
            graph = load_truth_data(data_path=args.data_path, dataset_name=data_name)
        else:
            graph = load_data(data_name)
            graph = inject_contextual_anomalies(graph=graph, k=K, p=P, q=Q_MAP[data_name], seed=seed)
            graph = inject_structural_anomalies(graph=graph, p=P, q=Q_MAP[data_name], seed=seed)

        label = graph.ndata['label']

        if args.model in ['RAND']:
            model = eval(f'{args.model}(**args_dict["model"])')
        else:
            raise ValueError(f"{args.model} is not implemented!")

        model.fit(graph, **args_dict["fit"])
        result = model.predict(graph, **args_dict["predict"])

        test_fpr, test_tpr = curve_obtain(label, result)
        final_score, _, _ = split_auc(label, result)
        the_ap_score = score_other(label, result)
        res_list_final.append(final_score)
        res_list_ap.append(the_ap_score)
        print(args_dict, flush=True)

    mean_final_auc = np.mean(np.array(res_list_final))
    std_final_auc = np.std(np.array(res_list_final))

    mean_final_ap = np.mean(np.array(res_list_ap))
    std_final_ap = np.std(np.array(res_list_ap))

    print("########### OVERALL AUC  #############", flush=True)
    print("final auc list: ", res_list_final, flush=True)
    print("final auc mean: ", mean_final_auc, flush=True)
    print("final auc std: ", std_final_auc, flush=True)
    print("&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&", flush=True)

    print("########### OTHER OVERALL PERFORMANCE  #############", flush=True)
    print("final ap list: ", res_list_ap, flush=True)
    print("final ap mean: ", mean_final_ap, flush=True)
    print("final ap std: ", std_final_ap, flush=True)
    print("&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&", flush=True)

    os._exit(0)