DataReader

class DLL.Data.DataReader(X, Y, batch_size=64, shuffle=True, shuffle_every_epoch=False)[source]

Bases: object

The data reader.

Parameters:

  • X (torch.Tensor of shape (n_samples, ...)) – The input data.

  • Y (torch.Tensor of shape (n_samples, ...)) – The target labels or values.

  • batch_size (int, optional) – The batch size. If the batch size is larger than the number of samples, the maximum number of samples is used as the batch_size. Must be a positive integer. Defaults to 64.

  • shuffle (bool, optional) – Determines if the input data is shuffled in the beginning. Defaults to True.

  • shuffle_every_epoch (bool, optional) – Determines if the input data is shuffled every time all data points are used. Defaults to False.

get_data()[source]

A generator giving going through the entire dataset.

Yields:

tuple[torch.Tensor, torch.Tensor] – X_batch, y_batch