def test(model, path):
##### put ur data_path of TestDataSet/Kvasir here #####
data_path = path
##### #####
model.eval()
image_root = '{}/images/'.format(data_path)
gt_root = '{}/masks/'.format(data_path)
test_loader = test_dataset(image_root, gt_root, 352)
b = 0.0
for i in range(100):
image, gt, name = test_loader.load_data()
gt = np.asarray(gt, np.float32)
gt /= (gt.max() + 1e-8)
image = image.cuda()
res = model(image)
res = F.upsample(res,
size=gt.shape,
mode='bilinear',
align_corners=False)
res = res.sigmoid().data.cpu().numpy().squeeze()
res = (res - res.min()) / (res.max() - res.min() + 1e-8)
input = res
target = np.array(gt)
N = gt.shape
smooth = 1
input_flat = np.reshape(input, (-1))
target_flat = np.reshape(target, (-1))
intersection = (input_flat * target_flat)
loss = (2 * intersection.sum() + smooth) / (input.sum() +
target.sum() + smooth)
a = '{:.4f}'.format(loss)
a = float(a)
b = b + a
return b / 100
for _data_name in [
'CVC-300', 'CVC-ClinicDB', 'Kvasir', 'CVC-ColonDB', 'ETIS-LaribPolypDB'
]:
data_path = './data/TestDataset/{}/'.format(_data_name)
save_path = './results/PraNet/{}/'.format(_data_name)
opt = parser.parse_args()
model = PraNet()
model.load_state_dict(torch.load(opt.pth_path))
model.cuda()
model.eval()
os.makedirs(save_path, exist_ok=True)
image_root = '{}/images/'.format(data_path)
gt_root = '{}/masks/'.format(data_path)
test_loader = test_dataset(image_root, gt_root, opt.testsize)
for i in range(test_loader.size):
image, gt, name = test_loader.load_data()
gt = np.asarray(gt, np.float32)
gt /= (gt.max() + 1e-8)
image = image.cuda()
res5, res4, res3, res2 = model(image)
res = res2
res = F.upsample(res,
size=gt.shape,
mode='bilinear',
align_corners=False)
res = res.sigmoid().data.cpu().numpy().squeeze()
res = (res - res.min()) / (res.max() - res.min() + 1e-8)
parser = argparse.ArgumentParser()
parser.add_argument('--testsize', type=int, default=352, help='testing size')
parser.add_argument('--pth_path', type=str, default='./snapshots/PraNet-ori.pth')
opt = parser.parse_args()
for _data_name in ['CVC-300', 'CVC-ClinicDB', 'CVC-ColonDB', 'ETIS-LaribPolypDB', 'Kvasir']:
data_path = './data/TestDataset/{}/'.format(_data_name)
save_path = './results/PraNet/{}/'.format(_data_name)
model = PraNet()
model.load(opt.pth_path)
model.eval()
os.makedirs(save_path, exist_ok=True)
image_root = '{}/images/'.format(data_path)
gt_root = '{}/masks/'.format(data_path)
test_loader = test_dataset(image_root, gt_root, opt.testsize) \
.set_attrs(batch_size=1, shuffle=False)
for image, gt, name in test_loader:
gt /= (gt.max() + 1e-08)
(res5, res4, res3, res2) = model(image)
res = res2
c, h, w = gt.shape
upsample = nn.upsample(res, size=(h, w), mode='bilinear')
res = res.sigmoid().data.squeeze()
res = ((res - res.min()) / ((res.max() - res.min()) + 1e-08))
print('> {} - {}'.format(_data_name, name))
imageio.imwrite((save_path + name[0]), res)