tumourkit.classification.train_graphs.train

tumourkit.classification.train_graphs.train(save_dir: str, save_weights: bool, tr_loader: GraphDataLoader, val_loader: GraphDataLoader, model: Module, optimizer: Optimizer, writer: SummaryWriter, n_early: int, device: str | None = 'cpu', check_iters: int | None = -1, conf: Dict[str, Any] | None = None, normalizers: Tuple[Normalizer] | None = None, num_classes: int | None = 2, enable_background: bool | None = False, use_neural_persistence: bool | None = False, use_cubical: bool | None = False) → None

Train the model with early stopping on either: F1 score (weighted) or neural persistence, or until 1000 iterations.

n_early is also called patience in some places.