paddlets.models.classify.dl.adapter.data_adapter

class ClassifyDataAdapter[source]

Bases: object

Data adapter, converts TSDataset to paddle.io.Dataset and paddle.io.DataLoader.

to_paddle_dataset(rawdatasets: List[TSDataset], labels: ndarray) → ClassifyPaddleDatasetImpl[source]

Converts TSDataset to paddle.io.Dataset.

Parameters

  • rawdataset (TSDataset) – Raw TSDataset for converting to paddle.io.Dataset.

  • labels – (np.ndarray) : The data class labels

Returns

A built PaddleDatasetImpl.

Return type

PaddleDatasetImpl

to_paddle_dataloader(paddle_dataset: ClassifyPaddleDatasetImpl, batch_size: int, collate_fn: Optional[Callable] = None, shuffle: bool = True) → DataLoader[source]

Converts paddle.io.Dataset to paddle.io.DataLoader.

Parameters

  • paddle_dataset (PaddleDatasetImpl) – Raw TSDataset for building paddle.io.DataLoader.

  • batch_size (int) – The number of samples for a single batch.

  • collate_fn (Callable, optional) – User-defined collate function for each batch, optional.

  • shuffle (bool, optional) – Whether to shuffle indices order before generating batch indices, default True.

Returns

A built paddle DataLoader.

Return type

PaddleDataLoader