Datasets

Dataset

Basic container for training data with labels

Parameters:

Name	Type	Description	Default
x	Tensor	training data values	None
y	Tensor	training data labels	None

split

Splits the data into training and validation Subsets

Parameters:

Name	Type	Description	Default
frac	float	The percent of data that will be for training	required
shuffle	bool	Whether the data will be shuffled before split. Defaults to True	True

Example

training, validation = dataset.split(frac=0.8, shuffle=True)

Dataloader

Iterates over a dataset in mini-batches.

The DataLoader handles batching and optional shuffling, providing an efficient interface for training models.

Parameters:

Name	Type	Description	Default
dataset	Dataset	Dataset to iterate over.	required
batch_size	int	Number of samples per batch.	None
shuffle	bool	Whether to shuffle the dataset at the beginning of each epoch. (defaults to False)	False
drop_remainder	bool	determins if the final mini batch should be dropped if the data cannot be segmented evenly. (defaults to False)	False

split

Splits the data into training and validation Dataloaders

Parameters:

Name	Type	Description	Default
frac	float	The percent of data that will be for training	required
shuffle	bool	Whether the data will be shuffled before split. Defaults to False	False

Example

training, validation = dataloader.split(frac=0.8, shuffle=False)