nebula.core.datasets.mnist.mnist

nebula.core.datasets.mnist.mnist#

Classes#

MNISTDataset

Abstract class for a partitioned dataset.

Module Contents#

class nebula.core.datasets.mnist.mnist.MNISTDataset(num_classes=10, partition_id=0, partitions_number=1, batch_size=32, num_workers=4, iid=True, partition='dirichlet', partition_parameter=0.5, seed=42, config=None)#

Bases: nebula.core.datasets.nebuladataset.NebulaDataset

Abstract class for a partitioned dataset.

Classes inheriting from this class need to implement specific methods for loading and partitioning the dataset.

initialize_dataset()#

Initialize the dataset. This should load or create the dataset.

load_mnist_dataset(train=True)#
generate_non_iid_map(dataset, partition='dirichlet', partition_parameter=0.5)#

Create a non-iid map of the dataset.

generate_iid_map(dataset, partition='balancediid', partition_parameter=2)#

Create an iid map of the dataset.

Contents