def load_data(self, data_path: str) -> list:
data_list = get_datalist(data_path)
t_data_list = []
# print("data_list:",data_list) .jpg - .txt
flag = 1
for img_path, label_path in data_list:
data = self._get_annotation(label_path)
# if flag:
# flag = 0
# print("data:",data)
if len(data['text_polys']) > 0:
item = {
'img_path': img_path,
'img_name': pathlib.Path(img_path).stem
}
# {'img_path': '/home/share/gaoluoluo/dbnet/datasets/train/img/ic15_img0.jpg', 'img_name': 'ic15_img0'}
item.update(data)
# if flag:
# flag = 0
# print("item:",item)
t_data_list.append(item)
else:
print('there is no suit bbox in {}'.format(label_path))
return t_data_list
def load_data(self, data_path: str) -> list:
data_list = get_datalist(data_path)
t_data_list = []
for img_path, label_path in data_list:
data = self._get_annotation(label_path)
if len(data['text_polys']) > 0:
item = {'img_path': img_path, 'img_name': pathlib.Path(img_path).stem}
item.update(data)
t_data_list.append(item)
else:
print('there is no suit bbox in {}'.format(label_path))
return t_data_list
def main():
opt = parse_opts()
opt.inputsize = [int(item) for item in opt.input_list.split(',')]
torch.cuda.set_device(opt.gpuid)
device = torch.device("cuda:%d" % opt.gpuid)
folder = 'fold_%d_%sresult' % (opt.fold, opt.model) + time.strftime(
"%Y_%m_%d%H_%M_%S", time.localtime())
save_path = os.path.join(opt.result_path, folder)
model = get_model(opt)
model.to(device)
trainlist, vallist = get_datalist(opt.fold)
trainset = DATASET(trainlist)
valset = DATASET_VAL(vallist)
train_loader = torch.utils.data.DataLoader(trainset,
batch_size=7,
num_workers=8,
shuffle=True)
val_loader = torch.utils.data.DataLoader(valset,
batch_size=2,
num_workers=6,
shuffle=False)
optimizer = RAdam(model.parameters(), lr=3e-4, weight_decay=1e-5)
best_meandice = 100
for epoch in range(50):
adjust_learning_rate(optimizer, epoch, 3e-4, 5)
train(model, optimizer, train_loader, epoch, 50, device)
mean_dist = val(model, val_loader, epoch, device)
if mean_dist < best_meandice:
best_meandice = mean_dist
check_mkdirs(save_path)
if epoch > 1:
print('save model...:' +
os.path.join(save_path, '%.4f.pkl' % (best_meandice)))
check_mkdirs(save_path)
torch.save(
model.state_dict(),
os.path.join(save_path, '%.4f.pkl' % (best_meandice)))
print('Best Mean Dice: %.4f' % best_meandice)
os.rename(save_path, save_path + '_%.4f' % best_meandice)
def load_data(self, data_path: str) -> list:
return get_datalist(data_path, self.num_label)
def load_data(self, data_path: str) -> list:
return get_datalist(data_path)
if checkValid:
try:
if not checkImageIsValid(imageBin):
print('%s is not a valid image' % imagePath)
continue
except:
continue
imageKey = 'image-%09d'.encode() % cnt
labelKey = 'label-%09d'.encode() % cnt
cache[imageKey] = imageBin
cache[labelKey] = label.encode()
if cnt % 1000 == 0:
writeCache(env, cache)
cache = {}
cnt += 1
nSamples = cnt - 1
cache['num-samples'.encode()] = str(nSamples).encode()
writeCache(env, cache)
print('Created dataset with %d samples' % nSamples)
if __name__ == '__main__':
data_list = [["/data1/ocr/dataset/test_crnn/train.txt"]]
save_path = '/data1/ocr/dataset/test_crn/lmdb/val'
os.makedirs(save_path, exist_ok=True)
train_data_list = get_datalist(data_list)
createDataset(train_data_list, save_path)