def create_coco_loader(*paths):
transform = utils.get_transform(config.image_size, config.central_fraction)
datasets = [data.CocoImages(path, transform=transform) for path in paths]
dataset = data.Composite(*datasets)
data_loader = torch.utils.data.DataLoader(
dataset,
batch_size=config.preprocess_batch_size,
num_workers=config.data_workers,
shuffle=False,
pin_memory=True,
)
return data_loader
def create_coco_loader(*paths):
transform = utils.get_transform(config.image_size, config.central_fraction)
datasets = [data.CocoImages(path, transform=transform) for path in paths]
#ipdb.set_trace() ## datasets[0].__getitem__(116591)[0] print the largest coco_id - this is within torch int64 bound!
dataset = data.Composite(*datasets)
data_loader = torch.utils.data.DataLoader(
dataset,
batch_size=config.preprocess_batch_size,
num_workers=config.data_workers,
shuffle=False,
pin_memory=True,
)
return data_loader
def create_data_loader(self, *paths):
""" Create a united PyTorch COCO data loader for every given path in the arguments"""
transform = utils.get_transform(self.image_size,
self.keep_central_fraction)
datasets = [
data.CocoImages(path, transform=transform) for path in paths
]
dataset = data.Composite(*datasets)
data_loader = torch.utils.data.DataLoader(
dataset,
batch_size=self.batch_size,
num_workers=self.num_threads_to_use,
shuffle=False,
pin_memory=True,
)
features_shape = (len(data_loader.dataset), config.output_features,
config.output_size, config.output_size)
return data_loader, features_shape
