-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtest.py
126 lines (103 loc) · 6.3 KB
/
test.py
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
import os
import argparse
import json
import torch.utils.data
from torch.utils.data import DataLoader
from tqdm import tqdm
import torch.backends.cudnn as cudnn
from torchvision.utils import save_image
from utils import prepare_sub_folder
from datasets import get_datasets
from models import create_model
import scipy.io as sio
import csv
import h5py
import numpy as np
parser = argparse.ArgumentParser(description='LVSPECT')
# model name
parser.add_argument('--experiment_name', type=str, default='experiment_train_casrec', help='give a experiment name before training')
parser.add_argument('--model_type', type=str, default='model_casrec', help='give a model name before training: model_svrhd / model_svrld / model_vm')
parser.add_argument('--resume', type=str, default=None, help='Filename of the checkpoint to resume')
# dataset
parser.add_argument('--data_root', type=str, default='../Data/Processed_Stress_LA_120degrees/', help='data root folder')
parser.add_argument('--dataset', type=str, default='LV', help='dataset name')
# network architectures, (discriminators e.g. cD, sD, are not used in the paper)
parser.add_argument('--net_G', type=str, default='scSERDUNet', help='generator network')
parser.add_argument('--nc', type=int, default=3, help='number of cascade')
parser.add_argument('--n_channels', type=int, default=32, help='number of input channels')
parser.add_argument('--UNet_depth', type=int, default=3, help='network depth')
parser.add_argument('--UNet_filters', type=int, default=5, help='UNet filters/channels in the first layer, 1 to 2^6')
parser.add_argument('--DuRDN_filters', type=int, default=64, help='DuRDN filters, 64')
parser.add_argument('--nConvlayers', type=int, default=4, help='Number of convolutional layers in each ResNet')
parser.add_argument('--growthRate', type=int, default=416, help='Growthrate')
parser.add_argument('--Consistency_Rate', type=float, default=0.01, help='data consistency rate')
parser.add_argument('--n_filters', type=int, default=32, help='ResNet, num of channels in RDB (inp or out)')
parser.add_argument('--growth_rate', type=int, default=32, help='ResNet, growth num of channels in each densely of RDB')
parser.add_argument('--n_denselayer', type=int, default=3, help='ResNet, num of dense layers in RDB')
# normalization
parser.add_argument('--norm', type=str, default='None', help='Normalization for each convolution') # 'BN' ,'IN', or 'None'
# training options
parser.add_argument('--n_epochs', type=int, default=1000, help='epoch')
parser.add_argument('--batch_size', type=int, default=16, help='training batch size')
parser.add_argument('--n_patch_train', type=int, default=1, help='number of patch to crop for training')
parser.add_argument('--patch_size_train', nargs='+', type=int, default=[60, 64, 64], help='randomly cropped patch size for train')
# evaluation options
parser.add_argument('--eval_epochs', type=int, default=4, help='evaluation epochs')
parser.add_argument('--save_epochs', type=int, default=4, help='save evaluation for every number of epochs')
parser.add_argument('--n_patch_test', type=int, default=1, help='number of patch to crop for evaluation')
parser.add_argument('--patch_size_test', nargs='+', type=int, default=[60, 64, 64], help='ordered cropped patch size for evaluation')
parser.add_argument('--n_patch_valid', type=int, default=1, help='number of patch to crop for evaluation')
parser.add_argument('--patch_size_valid', nargs='+', type=int, default=[60, 64, 64], help='ordered cropped patch size for evaluation')
# optimizer
parser.add_argument('--lr', type=float, default=1e-4, help='learning rate')
parser.add_argument('--lr_I1', type=float, default=1e-4, help='learning rate')
parser.add_argument('--lr_K1', type=float, default=1e-4, help='learning rate')
parser.add_argument('--lr_I2', type=float, default=1e-4, help='learning rate')
parser.add_argument('--lr_K2', type=float, default=1e-4, help='learning rate')
parser.add_argument('--beta1', type=float, default=0.5, help='beta1 for ADAM')
parser.add_argument('--beta2', type=float, default=0.999, help='beta2 for ADAM')
parser.add_argument('--weight_decay', type=float, default=0, help='weight decay')
# learning rate policy
parser.add_argument('--lr_policy', type=str, default='step', help='learning rate decay policy')
parser.add_argument('--step_size', type=int, default=1000, help='step size for step scheduler ')
parser.add_argument('--gamma', type=float, default=1, help='decay ratio for step scheduler')
# logger options
parser.add_argument('--snapshot_epochs', type=int, default=4, help='save model for every number of epochs')
parser.add_argument('--log_freq', type=int, default=100, help='save model for every number of epochs')
parser.add_argument('--output_path', default='./', type=str, help='Output path.')
# other
parser.add_argument('--num_workers', type=int, default=0, help='number of threads to load data')
parser.add_argument('--gpu_ids', type=int, nargs='+', default=[0], help='list of gpu ids')
opts = parser.parse_args()
options_str = json.dumps(opts.__dict__, indent=4, sort_keys=False)
print("------------------- Options -------------------")
print(options_str[2:-2])
print("-----------------------------------------------")
cudnn.benchmark = True
model = create_model(opts)
model.setgpu(opts.gpu_ids)
num_param = sum(p.numel() for p in model.parameters() if p.requires_grad)
print('Number of parameters: {} \n'.format(num_param))
if opts.resume is None:
model.initialize()
ep0 = -1
total_iter = 0
else:
ep0, total_iter = model.resume(opts.resume, train=False)
# select dataset
_, _, test_set = get_datasets(opts)
test_loader = DataLoader(dataset=test_set, num_workers=opts.num_workers, batch_size=1, shuffle=False)
# Setup directories
output_directory = os.path.join(opts.output_path, 'outputs', opts.experiment_name)
checkpoint_directory, image_directory = prepare_sub_folder(output_directory)
# evaluation
print('Normal Evaluation ......')
model.eval()
with torch.no_grad():
model.evaluate(test_loader)
model.save_images(test_loader, image_directory) # Save the image volume
# Record the epoch, psnr, ssim and mse
with open(os.path.join(image_directory, 'test_metrics_recon.csv'), 'w') as f: # Write CSV, some metadata
writer = csv.writer(f)
writer.writerow(['epoch', 'NMSE', 'NMAE', 'SSIM', 'PSNR'])
writer.writerow([ep0, model.nmse_recon, model.nmae_recon, model.ssim_recon, model.psnr_recon])