blueoil.datasets.cifar10 Module Contents Classes


Dataset base class

class blueoil.datasets.cifar10.Cifar10(subset='train', batch_size=100, *args, **kwargs)

Bases: blueoil.datasets.base.Base

Dataset base class

classes = ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']
extend_dir = CIFAR_10/cifar-10-batches-py
available_subsets = ['train', 'validation']
property num_per_epoch(self)

Returns the number of datas in the data subset.

_get_image(self, index)

Returns numpy array of an image

_unpickle(self, filename)
_load_data(self, filename)
__getitem__(self, i)

Returns the i-th item of the dataset.


returns the number of items in the dataset.