FedGDF/run_node_task_main.py at master · zzy211/FedGDF · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
from Node_level_Models.configs.config import args_parser
from Node_level_Models.helpers.metrics_utils import log_test_results

import numpy as np
import wandb
import os
import pandas as pd
# from node_clf import main as node_main
from time import time

args = args_parser()
rs = np.random.RandomState(args.seed)
seeds = rs.randint(1000, size=3)

project_name = [args.proj_name, args.proj_name+ "debug"]
proj_name = project_name[0]

def main(args):
    if args.alg_method == "FedTAD":
        from FedTAD_node_clf import main as node_main
    elif args.alg_method == "FedGEN":
        from FedGEN_node_clf import main as node_main
    elif args.alg_method == "Fedproto":
        from FedProto_node_clf import main as node_main
    elif args.alg_method == "FGPL":
        from FGPL_node_clf import main as node_main
    elif args.alg_method == "FedKD":    #our method
        from FedKD_node_clf import main as node_main
    elif args.alg_method == "MultiFedKD":
        from MultiFedKD_node_clf import main as node_main
    elif args.alg_method == "Local":
        from Local_node_clf import main as node_main
    elif args.alg_method == "FedKD_low_cost":
        from FedKD_low_cost_v3 import main as node_main
    elif args.alg_method == "FedTGP":
        from FedTGP_node_clf import main as node_main
    else:
        raise ValueError("alg_method is not defined!")

    model_name = args.model
    Alg_name = "Alg-" + args.alg_method
    file_name = Alg_name + 'Dataset-{}_Model-{}_IID-{}_Num_client-{}_Over_rate-{}'.format(
        args.dataset,
        model_name,
        args.is_iid,
        args.num_workers,
        args.overlapping_rate)

    average_overall_performance_list = []
    average_round_reach_target_list = []
    average_max_acc_list = []
    average_f1_score_list = []
    results_table = []
    metric_list = []

    for i in range(len(seeds)): #average the result of 5 tests
        args.seed = seeds[i]
        os.environ["WANDB_MODE"] = "offline" #offline or online
        # wandb init
        if args.alg_method == "BNS_GCN":
            log_dir = os.path.join(
            'wandb',
            args.model,
            args.dataset,
            f"alg_method_{args.alg_method}_ratio_node_{args.ratio_node}",
            f"num_workers_{args.num_workers}_dirichlet_alpha_{args.dirichlet_alpha}_round_{i}"
            )
        else:
            log_dir = os.path.join(
            'wandb',
            args.model,
            args.dataset,
            f"alg_method_{args.alg_method}",
            f"num_workers_{args.num_workers}_dirichlet_alpha_{args.dirichlet_alpha}_round_{i}"
            )
        os.makedirs(log_dir, exist_ok=True)
        logger = wandb.init(
            project=proj_name,
            group=file_name,
            name=f"round_{i}",
            config=args,
            dir=log_dir
        )

        average_overall_performance, round_reach_target_acc, max_acc, end_f1_score = node_main(args, logger)
        results_table.append([average_overall_performance, round_reach_target_acc, max_acc, end_f1_score])
        logger.log({"average_overall_performance": average_overall_performance, "round_reach_target": round_reach_target_acc, "max_acc": max_acc, "f1_score": end_f1_score})

        average_overall_performance_list.append(average_overall_performance)
        average_round_reach_target_list.append(round_reach_target_acc)
        average_max_acc_list.append(max_acc)
        average_f1_score_list.append(end_f1_score)

        # end the logger
        wandb.finish()

    # wandb table logger init
    columns = ["average_overall_performance_acc", "average_round_reach_target", "average_max_acc", "average_f1_score"]
    logger_table = wandb.Table(columns=columns, data=results_table)
    table_logger = wandb.init(
        #entity="hkust-gz",
        project=proj_name,
        group=file_name,
        name=f"exp_results",
        config=args,
    )
    table_logger.log({"results": logger_table})
    wandb.finish()

    mean_average_overall_performance, mean_average_round_reach_target, mean_max_accuracy, mean_end_f1_core = np.mean(np.array(average_overall_performance_list)), \
                                                                        np.mean(np.array(average_round_reach_target_list)),\
                                                                        np.mean(np.array(average_max_acc_list)),\
                                                                        np.mean(np.array(average_f1_score_list))

    std_average_overall_performance, std_average_round_reach_target, std_max_accuracy, std_end_f1_score = np.std(np.array(average_overall_performance_list)),\
                                                                      np.std(np.array(average_round_reach_target_list)),\
                                                                      np.std(np.array(average_max_acc_list)),\
                                                                      np.std(np.array(average_f1_score_list))

    header = ['dataset', 'model', 'method','num_workers', "mean_average_overall_performance", "std_average_overall_performance", "mean_average_round_reach_target", "std_average_round_reach_target", "mean_average_max_acc", "std_average_max_acc", "mean_average_f1_score", "std_average_f1_score"]
    paths = "./checkpoints/Node/"

    metric_list.append(args.dataset)
    metric_list.append(model_name)
    metric_list.append(args.alg_method)
    metric_list.append(args.num_workers)
    metric_list.append(mean_average_overall_performance)
    metric_list.append(std_average_overall_performance)
    metric_list.append(mean_average_round_reach_target)
    metric_list.append(std_average_round_reach_target)
    metric_list.append(mean_max_accuracy)
    metric_list.append(std_max_accuracy)
    metric_list.append(mean_end_f1_core)
    metric_list.append(std_end_f1_score)

    data = {
        "Metric": [
            "Dataset",
            "Model Name",
            'Method',
            'Num_Works',
            "Mean Avg Overall Performance",
            "Std Avg Overall Performance",
            "Mean Avg Round Reach Target",
            "Std Avg Round Reach Target",
            "Mean Avg Max Accuracy",
            "Std Avg Max Accuracy",
            "Mean Avg f1-Score",
            "Std Avg f1-Score"
        ],
        "Value": metric_list
    }

    df = pd.DataFrame(data)

    print(df)

if __name__ == '__main__':
    args = args_parser()
    main(args)