return lines
if __name__ == '__main__':
test_args = TestOptions().parse()
test_args.thread = 1
test_args.batchsize = 1
merge_cfg_from_file(test_args)
# load model
model = MetricDepthModel()
model.eval()
# load checkpoint
if test_args.load_ckpt:
load_ckpt(test_args, model)
model.cuda()
model = torch.nn.DataParallel(model)
out_dir = os.path.join(test_args.dataroot, 'VNL_Monocular')
if not os.path.exists(out_dir):
os.mkdir(out_dir)
file_path = test_args.dataroot + "path_list.txt"
with open(file_path, "r") as file:
for path in tqdm(file, total=get_num_lines(file_path)):
img_path = path.strip()
dir, file = os.path.split(img_path)
out_dir = dir.replace('leftImg8bit', 'VNL_Monocular')
out_path = os.path.join(
out_dir, file.replace('leftImg8bit', 'VNL_Monocular'))
optimizer = ModelOptimizer(model)
#loss function
loss_func = ModelLoss()
val_err = [{'abs_rel': 0, 'silog': 0}]
ignore_step = -1
# Lerning strategy
lr_optim_lambda = lambda iter: (1.0 - iter / (float(total_iters)))**0.9
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer.optimizer,
lr_lambda=lr_optim_lambda)
# load checkpoint
if train_args.load_ckpt:
load_ckpt(train_args, model, optimizer.optimizer, scheduler, val_err)
ignore_step = train_args.start_step - train_args.start_epoch * math.ceil(
train_datasize / train_args.batchsize)
if gpu_num != -1:
model = torch.nn.DataParallel(model)
try:
for epoch in range(train_args.start_epoch, cfg.TRAIN.EPOCH[-1]):
# training
train(train_dataloader, model, epoch, loss_func, optimizer,
scheduler, training_stats, val_dataloader, val_err,
ignore_step)
ignore_step = -1
except (RuntimeError, KeyboardInterrupt):
logger.info('Save ckpt on exception ...')