PlanetAlign.algorithms.PARROT

class PARROT(alpha: float = 0.5, rwr_restart_prob: float = 0.15, gamma: float = 0.75, lambda_p: float = 0.0005, lambda_e: float = 1e-05, lambda_n: float = 0.005, lambda_a: float = 0.0005, dtype: dtype = torch.float32)[source]

Bases: BaseModel

OT-based method PARROT for pairwise network alignment. PARROT is proposed by the paper “PARROT: Position-Aware Regularized Optimal Transport for Network Alignment.” in WWW 2023.

Parameters:

  • alpha (float, optional) – The hyparameter balancing the costs computed by RWR and node attributes. Default is 0.5.

  • rwr_restart_prob (float, optional) – The restart probability for the random walk with restart (RWR). Default is 0.15.

  • gamma (float, optional) – The discount factor of RWR. Default is 0.75.

  • lambda_p (float, optional) – The weight of the proximal point term. Default is 5e-4.

  • lambda_e (float, optional) – The weight of the edge consistency term. Default is 1e-5.

  • lambda_n (float, optional) – The weight of the neighborhood consistency term. Default is 5e-3.

  • lambda_a (float, optional) – The weight of the alignment preference term. Default is 5e-4.

  • dtype (torch.dtype, optional) – Data type of the tensors, choose from torch.float32 or torch.float64. Default is torch.float32.

test(dataset: Dataset, gids: List[int] | Tuple[int, ...], metrics: tuple[str] | list[str] = None)
Parameters:

  • dataset (Dataset) – The dataset containing the graphs to be aligned and the training/test data.

  • gids (list[int] or tuple[int, ...]) – The indices of the graphs in the dataset to be aligned.

  • metrics (tuple[str] or list[str], optional) – The metrics to be computed after alignment. Default is None, which computes Hits@K (K=1, 10, 30, 50) and MRR metrics.

train(dataset: Dataset, gid1: int, gid2: int, use_attr: bool = True, max_iters_sep_rwr: int = 100, max_iters_prod_rwr: int = 50, inner_iters: int = 5, outer_iters: int = 10, save_log: bool = True, verbose: bool = True)[source]
Parameters:

  • dataset (Dataset) – The dataset containing graphs to be aligned and the training/test data.

  • gid1 (int) – The graph id of the first graph to be aligned.

  • gid2 (int) – The graph id of the second graph to be aligned.

  • use_attr (bool, optional) – Flag for using attributes. Default is True.

  • max_iters_sep_rwr (int, optional) – Maximum number of iterations for separate RWR. Default is 100.

  • max_iters_prod_rwr (int, optional) – Maximum number of iterations for product RWR. Default is 50.

  • inner_iters (int, optional) – Number of inner iterations for the proximal point optimization. Default is 5.

  • outer_iters (int, optional) – Number of outer iterations for the proximal point optimization. Default is 10.

  • save_log (bool, optional) – Flag for saving the logs. Default is True.

  • verbose (bool, optional) – Flag for printing the logs. Default is True.