torchcell.losses.multi_dim_nan_tolerant.CombinedRegressionLoss

class CombinedRegressionLoss(loss_type: str = 'mse', weights: Tensor | None = None, quantile_spacing: float | None = None, huber_delta: float = 1.0)

Bases: Module

Dispatch to a NaN-tolerant regression loss and apply per-dimension weights.

forward(y_pred: Tensor, y_true: Tensor) → tuple[Tensor, Tensor]

Compute weighted loss while handling NaN values.