PAMIL/test.py at main · titizheng/PAMIL · 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
import os
os.environ['CUDA_VISIBLE_DEVICES'] = '7'
import sys


sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), "../..")))
from utilmodule.utils import make_parse
from utilmodule.core import test ,seed_torch
from torch.utils.data import DataLoader
from datasets.load_datasets import h5file_Dataset
import torch
import numpy as np
from utilmodule.createmode import create_model


def main(args):
    import pandas as pd

    seed_torch(2021)
    res_list = []

    basedmodel,ppo,classifymodel,memory,FusionHisF = create_model(args)
    data_csv_dir = args.csv
    h5file_dir = args.test_h5
<<<<<<< HEAD
    test_pth = ''
=======
    test_pth = '/home/ttzheng/MILRLMedical/PAMIL/save_model/Camelyon16_model.pth.tar'
>>>>>>> 9552a20 (Initial commit)

    checkpoint = torch.load(test_pth)
    basedmodel.load_state_dict(checkpoint['model_state_dict'])
    FusionHisF.load_state_dict(checkpoint['FusionHisF'])
    classifymodel.load_state_dict(checkpoint['fc'])
    ppo.policy.load_state_dict(checkpoint['policy'])

    basedmodel.eval()
    classifymodel.eval()
    FusionHisF.eval()
    ppo.policy.eval()

    test_dataset = h5file_Dataset(data_csv_dir,h5file_dir,'test')
    test_dataloader = DataLoader(test_dataset, batch_size=1, shuffle=False)

    precision, recall, f1, auc, accuracy = test(args,basedmodel,ppo,classifymodel,FusionHisF,memory,test_dataloader )
    res_list.append([accuracy,auc,precision,recall,f1])

    df = pd.DataFrame(res_list, columns=[ 'acc', 'auc', 'precision', 'recall', 'f1'])

<<<<<<< HEAD
    df.to_csv('/result.csv', index=False)
=======
    df.to_csv('/home/ttzheng/MILRLMedical/PAMIL/save_model/result.csv', index=False)
>>>>>>> 9552a20 (Initial commit)


if __name__ == "__main__":

    args = make_parse()
    main(args)