2020-09-17 19:30:32 +00:00
|
|
|
import os.path as osp
|
|
|
|
import logging
|
|
|
|
import time
|
|
|
|
import argparse
|
|
|
|
|
|
|
|
import os
|
2021-06-25 19:16:15 +00:00
|
|
|
|
|
|
|
from torchvision.transforms import CenterCrop
|
|
|
|
|
|
|
|
from trainer.ExtensibleTrainer import ExtensibleTrainer
|
2020-10-14 02:56:39 +00:00
|
|
|
from utils import options as option
|
2020-09-17 19:30:32 +00:00
|
|
|
import utils.util as util
|
|
|
|
from data import create_dataset, create_dataloader
|
|
|
|
from tqdm import tqdm
|
|
|
|
import torch
|
|
|
|
import torchvision
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
2021-06-25 19:16:15 +00:00
|
|
|
bin_path = "f:\\tmp\\binned"
|
|
|
|
good_path = "f:\\tmp\\good"
|
2020-09-17 19:30:32 +00:00
|
|
|
os.makedirs(bin_path, exist_ok=True)
|
|
|
|
os.makedirs(good_path, exist_ok=True)
|
|
|
|
|
|
|
|
|
|
|
|
torch.backends.cudnn.benchmark = True
|
|
|
|
parser = argparse.ArgumentParser()
|
2021-06-25 19:16:15 +00:00
|
|
|
parser.add_argument('-opt', type=str, help='Path to options YAML file.', default='../options/train_quality_detectors/train_resnet_jpeg.yml')
|
2020-09-17 19:30:32 +00:00
|
|
|
opt = option.parse(parser.parse_args().opt, is_train=False)
|
|
|
|
opt = option.dict_to_nonedict(opt)
|
|
|
|
opt['dist'] = False
|
|
|
|
|
|
|
|
util.mkdirs(
|
|
|
|
(path for key, path in opt['path'].items()
|
|
|
|
if not key == 'experiments_root' and 'pretrain_model' not in key and 'resume' not in key))
|
|
|
|
util.setup_logger('base', opt['path']['log'], 'test_' + opt['name'], level=logging.INFO,
|
|
|
|
screen=True, tofile=True)
|
|
|
|
logger = logging.getLogger('base')
|
|
|
|
logger.info(option.dict2str(opt))
|
|
|
|
|
|
|
|
#### Create test dataset and dataloader
|
|
|
|
test_loaders = []
|
|
|
|
for phase, dataset_opt in sorted(opt['datasets'].items()):
|
|
|
|
test_set = create_dataset(dataset_opt)
|
|
|
|
test_loader = create_dataloader(test_set, dataset_opt, opt=opt)
|
|
|
|
logger.info('Number of test images in [{:s}]: {:d}'.format(dataset_opt['name'], len(test_set)))
|
|
|
|
test_loaders.append(test_loader)
|
|
|
|
|
2021-06-25 19:16:15 +00:00
|
|
|
model = ExtensibleTrainer(opt)
|
2020-09-17 19:30:32 +00:00
|
|
|
fea_loss = 0
|
|
|
|
for test_loader in test_loaders:
|
|
|
|
test_set_name = test_loader.dataset.opt['name']
|
|
|
|
logger.info('\nTesting [{:s}]...'.format(test_set_name))
|
|
|
|
test_start_time = time.time()
|
|
|
|
dataset_dir = osp.join(opt['path']['results_root'], test_set_name)
|
|
|
|
util.mkdir(dataset_dir)
|
|
|
|
|
|
|
|
tq = tqdm(test_loader)
|
2020-09-26 04:19:38 +00:00
|
|
|
removed = 0
|
2020-10-10 01:21:43 +00:00
|
|
|
means = []
|
2021-06-25 19:16:15 +00:00
|
|
|
for k, data in enumerate(tq):
|
|
|
|
model.feed_data(data, k)
|
2020-09-17 19:30:32 +00:00
|
|
|
model.test()
|
2021-06-25 19:16:15 +00:00
|
|
|
results = torch.argmax(torch.nn.functional.softmax(model.eval_state['logits'][0], dim=-1), dim=1)
|
2020-09-17 19:30:32 +00:00
|
|
|
for i in range(results.shape[0]):
|
2021-06-25 19:16:15 +00:00
|
|
|
if results[i] == 0:
|
|
|
|
imname = osp.basename(data['HQ_path'][i])
|
|
|
|
# For VERIFICATION:
|
|
|
|
#torchvision.utils.save_image(data['hq'][i], osp.join(bin_path, imname))
|
|
|
|
# 4 REALZ:
|
|
|
|
os.remove(data['HQ_path'][i])
|
|
|
|
removed += 1
|
2020-09-17 19:30:32 +00:00
|
|
|
|
2020-09-26 04:19:38 +00:00
|
|
|
print("Removed %i/%i images" % (removed, len(test_set)))
|