def __init__(self, dataset="nyu"):
print("VNL init")
self.dataset = dataset
if dataset == "nyu":
test_args = Namespace(
batchsize=2,
cfg_file='lib/configs/resnext101_32x4d_nyudv2_class',
dataroot='./',
dataset='any',
epoch=30,
load_ckpt='./nyu_rawdata.pth',
phase='test',
phase_anno='test',
results_dir='./evaluation',
resume=False,
start_epoch=0,
start_step=0,
thread=4,
use_tfboard=False)
elif dataset == "kitti":
test_args = Namespace(
batchsize=2,
cfg_file='lib/configs/resnext101_32x4d_kitti_class',
dataroot='./',
dataset='kitti',
epoch=30,
load_ckpt='./kitti_eigen.pth',
phase='test',
phase_anno='test',
results_dir='./evaluation',
resume=False,
start_epoch=0,
start_step=0,
thread=4,
use_tfboard=False)
test_args.thread = 1
test_args.batchsize = 1
merge_cfg_from_file(test_args)
self.model = MetricDepthModel()
self.model.eval()
# load checkpoint
if dataset == "nyu":
load_ckpt("./vnl/nyu_rawdata.pth", self.model)
elif dataset == "kitti":
load_ckpt("./vnl/kitti_eigen.pth", self.model)
self.model.cuda()
self.model = torch.nn.DataParallel(self.model)
def get_num_lines(file_path):
fp = open(file_path, "r+")
buf = mmap.mmap(fp.fileno(), 0)
lines = 0
while buf.readline():
lines += 1
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)
# Train args
train_opt = TrainOptions()
train_args = train_opt.parse()
train_opt.print_options(train_args)
# Validation args
val_opt = ValOptions()
val_args = val_opt.parse()
val_args.batchsize = 1
val_args.thread = 0
val_opt.print_options(val_args)
train_dataloader = CustomerDataLoader(train_args)
train_datasize = len(train_dataloader)
gpu_num = torch.cuda.device_count()
merge_cfg_from_file(train_datasize, gpu_num)
val_dataloader = CustomerDataLoader(val_args)
val_datasize = len(val_dataloader)
# Print configs
print_configs(cfg)
# tensorboard logger
if train_args.use_tfboard:
from tensorboardX import SummaryWriter
tblogger = SummaryWriter(cfg.TRAIN.LOG_DIR)
# training status for logging
training_stats = TrainingStats(
train_args, cfg.TRAIN.LOG_INTERVAL,
# Train args
train_opt = TrainOptions()
train_args = train_opt.parse()
train_opt.print_options(train_args)
# Validation args
val_opt = ValOptions()
val_args = val_opt.parse()
val_args.batchsize = 1
val_args.thread = 0
val_opt.print_options(val_args)
train_dataloader = CustomerDataLoader(train_args)
train_datasize = len(train_dataloader)
gpu_num = torch.cuda.device_count()
merge_cfg_from_file(train_args)
val_dataloader = CustomerDataLoader(val_args)
val_datasize = len(val_dataloader)
# Print configs
print_configs(cfg)
# tensorboard logger
if train_args.use_tfboard:
from tensorboardX import SummaryWriter
tblogger = SummaryWriter(cfg.TRAIN.LOG_DIR)
# training status for logging
training_stats = TrainingStats(
train_args, cfg.TRAIN.LOG_INTERVAL,