cora

Bases: graph_dataloader

A dataloader class for the Cora dataset.

This class extends graph_dataloader to handle the Cora graph dataset, which is commonly used in graph-based machine learning research.

Attributes:

Name	Type	Description
data_profile	dict	Metadata and download links specific to the Cora dataset.
graph	graph	The loaded graph structure for the Cora dataset.

Methods:

Name	Description
__init__	Initializes the dataloader for the Cora dataset.
get_train_test_idx	Generates train and test indices for the Cora dataset.

Source code in tinybig/data/graph_dataloader.py

class cora(graph_dataloader):
    """
    A dataloader class for the Cora dataset.

    This class extends `graph_dataloader` to handle the Cora graph dataset,
    which is commonly used in graph-based machine learning research.

    Attributes
    ----------
    data_profile: dict
        Metadata and download links specific to the Cora dataset.
    graph: graph_class
        The loaded graph structure for the Cora dataset.

    Methods
    -------
    __init__(name: str = 'cora', train_batch_size: int = 64, test_batch_size: int = 64, ...)
        Initializes the dataloader for the Cora dataset.
    get_train_test_idx(X: torch.Tensor = None, y: torch.Tensor = None, ...)
        Generates train and test indices for the Cora dataset.
    """
    def __init__(self, name: str = 'cora', train_batch_size: int = 64, test_batch_size: int = 64, *args, **kwargs):
        """
        Initializes the dataloader for the Cora dataset.

        Parameters
        ----------
        name: str, default = 'cora'
            The name of the dataset.
        train_batch_size: int, default = 64
            Batch size for the training dataset.
        test_batch_size: int, default = 64
            Batch size for the testing dataset.

        Returns
        -------
        None
        """
        super().__init__(data_profile=CORA_DATA_PROFILE, name=name, train_batch_size=train_batch_size, test_batch_size=test_batch_size)

    def get_train_test_idx(self, X: torch.Tensor = None, y: torch.Tensor = None, *args, **kwargs):
        """
        Generates train and test indices for the Cora dataset.

        Parameters
        ----------
        X: torch.Tensor, optional
            Node features (not used in this method).
        y: torch.Tensor, optional
            Labels (not used in this method).

        Returns
        -------
        tuple
            Train indices (`torch.LongTensor`) and test indices (`torch.LongTensor`).

        Notes
        -----
        The train indices are predefined as the first 140 nodes.
        The test indices are predefined as nodes 500 to 1499.
        """
        train_idx = torch.LongTensor(range(140))
        test_idx = torch.LongTensor(range(500, 1500))
        return train_idx, test_idx

__init__(name='cora', train_batch_size=64, test_batch_size=64, *args, **kwargs)

Initializes the dataloader for the Cora dataset.

Parameters:

Name	Type	Description	Default
name	str	The name of the dataset.	'cora'
train_batch_size	int	Batch size for the training dataset.	64
test_batch_size	int	Batch size for the testing dataset.	64

Returns:

Type	Description
None

Source code in tinybig/data/graph_dataloader.py

def __init__(self, name: str = 'cora', train_batch_size: int = 64, test_batch_size: int = 64, *args, **kwargs):
    """
    Initializes the dataloader for the Cora dataset.

    Parameters
    ----------
    name: str, default = 'cora'
        The name of the dataset.
    train_batch_size: int, default = 64
        Batch size for the training dataset.
    test_batch_size: int, default = 64
        Batch size for the testing dataset.

    Returns
    -------
    None
    """
    super().__init__(data_profile=CORA_DATA_PROFILE, name=name, train_batch_size=train_batch_size, test_batch_size=test_batch_size)

get_train_test_idx(X=None, y=None, *args, **kwargs)

Generates train and test indices for the Cora dataset.

Parameters:

Name	Type	Description	Default
X	Tensor	Node features (not used in this method).	None
y	Tensor	Labels (not used in this method).	None

Returns:

Type	Description
tuple	Train indices (torch.LongTensor) and test indices (torch.LongTensor).

Notes

The train indices are predefined as the first 140 nodes. The test indices are predefined as nodes 500 to 1499.

Source code in tinybig/data/graph_dataloader.py

def get_train_test_idx(self, X: torch.Tensor = None, y: torch.Tensor = None, *args, **kwargs):
    """
    Generates train and test indices for the Cora dataset.

    Parameters
    ----------
    X: torch.Tensor, optional
        Node features (not used in this method).
    y: torch.Tensor, optional
        Labels (not used in this method).

    Returns
    -------
    tuple
        Train indices (`torch.LongTensor`) and test indices (`torch.LongTensor`).

    Notes
    -----
    The train indices are predefined as the first 140 nodes.
    The test indices are predefined as nodes 500 to 1499.
    """
    train_idx = torch.LongTensor(range(140))
    test_idx = torch.LongTensor(range(500, 1500))
    return train_idx, test_idx