def train_factory(args, preprocess, target_transforms):
train_datas = [datasets.CocoKeypoints(
root=args.train_image_dir,
annFile=item,
preprocess=preprocess,
image_transform=transforms.image_transform_train,
target_transforms=target_transforms,
n_images=args.n_images,
) for item in args.train_annotations]
train_data = torch.utils.data.ConcatDataset(train_datas)
train_loader = torch.utils.data.DataLoader(
train_data, batch_size=args.batch_size, shuffle=True,
pin_memory=args.pin_memory, num_workers=args.loader_workers, drop_last=True)
val_data = datasets.CocoKeypoints(
root=args.val_image_dir,
annFile=args.val_annotations,
preprocess=preprocess,
image_transform=transforms.image_transform_train,
target_transforms=target_transforms,
n_images=args.n_images,
)
val_loader = torch.utils.data.DataLoader(
val_data, batch_size=args.batch_size, shuffle=False,
pin_memory=args.pin_memory, num_workers=args.loader_workers, drop_last=True)
return train_loader, val_loader, train_data, val_data
def test_dataloader(self):
val_data = datasets.CocoKeypoints(
root=self.args.val_image_dir,
annFile=self.args.val_annotations,
preprocess=preprocess,
image_transform=transforms.image_transform_train,
target_transforms=self.target_transforms,
n_images=self.args.n_images,
)
val_loader = torch.utils.data.DataLoader(
val_data, batch_size=self.args.batch_size, shuffle=False,
pin_memory=self.args.pin_memory, num_workers=self.args.loader_workers, drop_last=True)
return val_loader
def test_dataloader(self):
val_data = datasets.CocoKeypoints(
root=cfg.DATASET.VAL_IMAGE_DIR,
annFile=cfg.DATASET.VAL_ANNOTATIONS,
preprocess=preprocess,
image_transform=transforms.image_transform_train,
target_transforms=self.target_transforms,
n_images=None,
)
val_loader = torch.utils.data.DataLoader(
val_data, batch_size=cfg.TEST.BATCH_SIZE_PER_GPU*len(cfg.GPUS), shuffle=False,
pin_memory=cfg.PIN_MEMORY, num_workers=cfg.WORKERS, drop_last=True)
return val_loader
def tng_dataloader(self):
train_datas = [datasets.CocoKeypoints(
root=self.args.train_image_dir,
annFile=item,
preprocess=preprocess,
image_transform=transforms.image_transform_train,
target_transforms=self.target_transforms,
n_images=args.n_images,
) for item in self.args.train_annotations]
train_data = torch.utils.data.ConcatDataset(train_datas)
train_loader = torch.utils.data.DataLoader(
train_data, batch_size=self.args.batch_size, shuffle=True,
pin_memory=self.args.pin_memory, num_workers=self.args.loader_workers, drop_last=True)
return train_loader
def tng_dataloader(self):
train_datas = [datasets.CocoKeypoints(
root=cfg.DATASET.TRAIN_IMAGE_DIR,
annFile=item,
preprocess=preprocess,
image_transform=transforms.image_transform_train,
target_transforms=self.target_transforms,
n_images=None,
) for item in cfg.DATASET.TRAIN_ANNOTATIONS]
train_data = torch.utils.data.ConcatDataset(train_datas)
train_loader = torch.utils.data.DataLoader(
train_data, batch_size=cfg.TRAIN.BATCH_SIZE_PER_GPU*len(cfg.GPUS), shuffle=True,
pin_memory=cfg.PIN_MEMORY, num_workers=cfg.WORKERS, drop_last=True)
return train_loader
stds = [0.229, 0.224, 0.225]
image = image.transpose((1,2,0))
for i in range(3):
image[:, :, i] = image[:, :, i] * stds[i]
image[:, :, i] = image[:, :, i] + means[i]
image = image.copy()[:,:,::-1]
image = image*255
return image
train_datas = [datasets.CocoKeypoints(
root=cfg.DATASET.TRAIN_IMAGE_DIR,
annFile=item,
preprocess=preprocess,
image_transform=transforms.image_transform_train,
target_transforms=None,
n_images=None,
) for item in cfg.DATASET.TRAIN_ANNOTATIONS]
train_data = torch.utils.data.ConcatDataset(train_datas)
train_loader = torch.utils.data.DataLoader(
train_data, batch_size=cfg.TRAIN.BATCH_SIZE_PER_GPU*len(cfg.GPUS), shuffle=True,
pin_memory=cfg.PIN_MEMORY, num_workers=cfg.WORKERS, drop_last=True)
val_data = datasets.CocoKeypoints(
root=cfg.DATASET.VAL_IMAGE_DIR,
annFile=cfg.DATASET.VAL_ANNOTATIONS,
preprocess=preprocess,
