data.diabetes

data.diabetes

Classes

Name	Description
DiabetesDataset	Diabetes dataset wrapping sklearn’s diabetes dataset or custom data.

DiabetesDataset

data.diabetes.DiabetesDataset(
    X=None,
    y=None,
    transform=None,
    target_transform=None,
)

Diabetes dataset wrapping sklearn’s diabetes dataset or custom data.

Functions

Name	Description
get_diabetes_dataloaders	Returns train and test dataloaders for the Diabetes dataset.

get_diabetes_dataloaders

data.diabetes.get_diabetes_dataloaders(
    test_size=0.2,
    batch_size=32,
    scale_features=True,
    shuffle_train=True,
    shuffle_test=False,
    random_state=42,
    num_workers=0,
    pin_memory=False,
)

Returns train and test dataloaders for the Diabetes dataset.

Parameters

Name	Type	Description	Default
test_size	float	Fraction of data to use for testing.	`0.2`
batch_size	int	Batch size.	`32`
scale_features	bool	Whether to standardize features using StandardScaler.	`True`
shuffle_train	bool	Whether to shuffle the training data.	`True`
shuffle_test	bool	Whether to shuffle the test data.	`False`
random_state	int	Random seed for splitting.	`42`
num_workers	int	Number of subprocesses to use for data loading.	`0`
pin_memory	bool	If True, the data loader will copy Tensors into CUDA pinned memory before returning them.	`False`

Returns

Name	Type	Description
tuple	Tuple[`DataLoader`, `DataLoader`, Optional[`StandardScaler`]]	(train_loader, test_loader, scaler) scaler is the StandardScaler implementation if scale_features=True, else None.

Examples

from spotoptim.data.diabetes import get_diabetes_dataloaders

train_loader, test_loader, scaler = get_diabetes_dataloaders()
print(train_loader)
print(test_loader)
print(scaler)

<torch.utils.data.dataloader.DataLoader object at 0x11f80c2f0>
<torch.utils.data.dataloader.DataLoader object at 0x11f7720d0>
StandardScaler()