build_mnist_dataloader#
- composer.datasets.build_mnist_dataloader(datadir, global_batch_size, is_train=True, download=True, drop_last=True, shuffle=True, **dataloader_kwargs)[source]#
Builds an MNIST dataloader.
- Parameters
datadir (str) โ Path to the data directory
global_batch_size (int) โ Global batch size.
is_train (bool) โ Whether to load the training data or validation data. Default:
True
.download (bool, optional) โ Whether to download the dataset, if needed. Default:
True
.drop_last (bool) โ Drop remainder samples. Default:
True
.shuffle (bool) โ Shuffle the dataset. Default:
True
.**dataloader_kwargs (Any) โ Additional settings for the dataloader (e.g. num_workers, etc.)