###############################################################################
# Load train data
###############################################################################
PPT = [cfg.PROJ.TRAIN_PPT, (cfg.PROJ.TRAIN_PPT + cfg.PROJ.EVAL_PPT)]
print(f"{gct()} : Loading traning data")
train_data = DataLoader(
HpatchDataset(
data_type="train",
PPT=PPT,
use_all=cfg.PROJ.TRAIN_ALL,
csv_file=cfg[cfg.PROJ.TRAIN]["csv"],
root_dir=cfg[cfg.PROJ.TRAIN]["root"],
transform=transforms.Compose([
Grayscale(),
Normalize(mean=cfg[cfg.PROJ.TRAIN]["MEAN"],
std=cfg[cfg.PROJ.TRAIN]["STD"]),
LargerRescale((960, 1280)),
RandomCrop((720, 960)),
Rescale((240, 320)),
ToTensor(),
]),
),
batch_size=cfg.TRAIN.BATCH_SIZE,
shuffle=True,
num_workers=0,
)
###############################################################################
# Load evaluation data
###############################################################################
root_dir += 'EFDataset'
seq = 'ef'
a = True
else:
print(f'cannot find {args.data}')
exit(-1)
mean = cfg[seq]["MEAN"]
std = cfg[seq]["STD"]
data_loader = DataLoader(HpatchDataset(
data_type="test",
PPT=[0.8, 0.9],
use_all=a,
csv_file=csv_file,
root_dir=root_dir,
transform=transforms.Compose([
Grayscale(),
Normalize(mean=mean, std=std),
Rescale((960, 1280)),
Rescale((480, 640)),
ToTensor()
]),
),
batch_size=1,
shuffle=False,
num_workers=0)
useful_list = []
repeat_list = []
with torch.no_grad():
for i_batch, sample_batched in enumerate(data_loader, 1):
im1_data, im1_info, homo12, im2_data, im2_info, homo21, im1_raw, im2_raw = parse_batch(
a = True
else:
print(f'cannot find {args.data}')
exit(-1)
mean=cfg[seq]["MEAN"]
std=cfg[seq]["STD"]
data_loader = DataLoader(
HpatchDataset(
data_type="test",
PPT=0.9,
use_all=a,
csv_file=csv_file,
root_dir=root_dir,
transform=transforms.Compose(
[
Grayscale(),
Normalize(mean=mean, std=std),
Rescale((960, 1280)),
Rescale((240, 320)),
ToTensor()
]
),
),
batch_size=1,
shuffle=False,
num_workers=0
)
useful_list = []
repeat_list = []
for i_batch, sample_batched in enumerate(data_loader, 1):