o }o™iB ã@s‚ddlmZddlmZddlmZddlmZmZm Z ddgZ Gdd„dej eeƒZ Gdd„dej eeƒZ eGd d „d ƒƒZ d S) é)Ú dataclass)ÚOptional)Údata)Ú SerializationÚTypingÚ typecheckÚDatasetÚIterableDatasetc@ó&eZdZdZdd„Zeƒdd„ƒZdS)rzrDataset with output ports Please Note: Subclasses of IterableDataset should *not* implement input_types. cCó tj |¡S©zŒ A default implementation of a collation function. Users should override this method to define custom data loaders. ©rÚ dataloaderÚdefault_collate©ÚselfÚbatch©rúM/home/ubuntu/.local/lib/python3.10/site-packages/nemo/core/classes/dataset.pyÚ _collate_fnó zDataset._collate_fncCó|jdur tdƒ‚| |¡S)a This is the method that user pass as functor to DataLoader. The method optionally performs neural type checking and add types to the outputs. Please note, subclasses of Dataset should not implement `input_types`. Usage: .. code-block:: python dataloader = torch.utils.data.DataLoader( ...., collate_fn=dataset.collate_fn, .... ) Returns: Collated batch, with or without types. NúCDatasets should not implement `input_types` as they are not checked©Ú input_typesÚ TypeErrorrrrrrÚ collate_fn%s  zDataset.collate_fnN©Ú__name__Ú __module__Ú __qualname__Ú__doc__rrrrrrrró c@r )r z{Iterable Dataset with output ports Please Note: Subclasses of IterableDataset should *not* implement input_types. cCr r r rrrrrGrzIterableDataset._collate_fncCr)a” This is the method that user pass as functor to DataLoader. The method optionally performs neural type checking and add types to the outputs. # Usage: dataloader = torch.utils.data.DataLoader( ...., collate_fn=dataset.collate_fn, .... ) Returns: Collated batch, with or without types. NrrrrrrrNs  zIterableDataset.collate_fnNrrrrrr Ar"c@sNeZdZUdZeed<dZeed<dZeed<dZ e eed<dZ eed <d S) Ú DatasetConfigé Ú batch_sizeFÚ drop_lastÚshufflerÚ num_workersTÚ pin_memoryN) rrr r%ÚintÚ__annotations__r&Úboolr'r(rr)rrrrr#es   r#N)Ú dataclassesrÚtypingrÚ torch.utilsrÚnemo.core.classesrrrÚ__all__rr r#rrrrÚs  )$