EllipticDataset(root, name, transform=None, pre_transform=None, pre_filter=None)

Bases: InMemoryDataset

Elliptic Bitcoin transaction network dataset loader for graph-based analysis.

Parameters:
  • root (str) –

    Root directory where the dataset should be saved

  • name (str) –

    Name of the elliptic dataset (must be in format 'tx{N}' where N is 0-48)

  • transform (callable, default: None ) –

    Function/transform that takes in a Data object and returns a transformed version. Default: None

  • pre_transform (callable, default: None ) –

    Function/transform to be applied to the data object before saving. Default: None

  • pre_filter (callable, default: None ) –

    Function that takes in a Data object and returns a boolean value, indicating whether the data object should be included. Default: None
Notes

Dataset Structure:

  • Nodes represent Bitcoin transactions
  • Edges represent transaction flows
  • Node features from transaction data
  • Labels indicate transaction categories
  • Includes train/val/test splits (80/10/10)
processed_file_names property

Names of processed data files.

Returns:
  • list[str] –

    List of processed file names
Notes

Processed files:

  • data.pt: Contains processed PyTorch Geometric data object
raw_file_names property

Names of required raw files.

Returns:
  • list[str] –

    List of required raw file names
Notes

Required files:

  • data.pkl: Pickle file containing adjacency matrix, labels, and features
download()

Download raw data files.

Notes

Empty implementation - data should be manually placed in raw directory

load_data(data_dir, filename)

Load raw pickle dataset file.

Parameters:
  • data_dir (str) –

    Path to the pickle file

  • filename (str) –

    Name of the dataset file (must be in format 'tx{N}' where N is 0-48)
Returns:
  • tuple[Tensor, Tensor, Tensor] –

    Contains:

    • x: Node feature matrix [num_nodes, num_features]
    • edge_index: Graph connectivity [2, num_edges]
    • y: Node labels [num_nodes]
Notes
  • Validates dataset number
  • Loads pickle file containing network structure
  • Converts data to PyTorch tensors
process(mask=True)

Process raw data into PyTorch Geometric Data format.

Parameters:
  • mask (bool, default: True ) –

    Whether to generate train/val/test masks. Default: True
Notes

Processing Steps:

  • Load pickle data:

    • Node features
    • Adjacency matrix
    • Node labels

  • Convert to PyTorch format:

    • Edge indices from sparse adjacency
    • Float features
    • Integer labels

  • Create Data object with:

    • Edge indices
    • Node features
    • Node labels
    • Train/val/test masks

  • Apply pre-transform if specified

  • Save processed data

Data Split:

  • Training: 80%
  • Validation: 10%
  • Testing: 10%

Features:

  • Sparse matrix conversion
  • Type casting
  • Random split generation
  • Optional pre-transform support