CIFAR-100 Data Set¶
-
class
deepobs.pytorch.datasets.cifar100.cifar100(batch_size, data_augmentation=True, train_eval_size=10000)[source]¶ DeepOBS data set class for the CIFAR-100 data set.
Parameters: - batch_size (int) -- The mini-batch size to use. Note that, if
batch_sizeis not a divider of the dataset size (50 000for train,10 000for test) the remainder is dropped in each epoch (after shuffling). - data_augmentation (bool) -- If
Truesome data augmentation operations (random crop window, horizontal flipping, lighting augmentation) are applied to the training data (but not the test data). - train_eval_size (int) -- Size of the train eval data set.
Defaults to
10 000the size of the test set.
-
_make_dataloader()¶ A helper that is shared by all three data loader methods.
- batch_size (int) -- The mini-batch size to use. Note that, if