gnn_tracking.models.noise_classification#

Models for filtering out noise before we even build a graph

Classes#

TruthNoiseClassifierModel

Remove all noise with truth information

WithNoiseClassification

Combine a noise filter with another model

Functions#

obj_from_or_to_hparams(→ Any)

Used to support initializing python objects from hyperparameters:

Module Contents#

gnn_tracking.models.noise_classification.obj_from_or_to_hparams(self: pytorch_lightning.core.mixins.hparams_mixin.HyperparametersMixin, key: str, obj: Any) Any#

Used to support initializing python objects from hyperparameters: If obj is a python object other than a dictionary, its hyperparameters are saved (its class path and init args) to self.hparams[key]. If obj is instead a dictionary, its assumed that we have to restore an object based on this information.

class gnn_tracking.models.noise_classification.TruthNoiseClassifierModel#

Bases: torch.nn.Module

Remove all noise with truth information

forward(data: torch_geometric.data.Data) torch_geometric.data.Data#
class gnn_tracking.models.noise_classification.WithNoiseClassification(noise_model: torch.nn.Module, model: torch.nn.Module)#

Bases: torch.nn.Module, pytorch_lightning.core.mixins.hparams_mixin.HyperparametersMixin

Combine a noise filter with another model

forward(data: torch_geometric.data.Data) dict[str, torch.Tensor | None]#