beta_nll_loss#
- radionets.core.loss_functions.beta_nll_loss(x: tensor, y: tensor, beta: float = 0.5)[source]#
Compute beta-NLL loss
- Parameters:
- x
torch.tensor() Prediction of the model.
- y
torch.tensor() Ground truth.
- betafloat
Parameter from range [0, 1] controlling relative weighting between data points, where “0” corresponds to high weight on low error points and “1” to an equal weighting.
- x
- Returns:
- floatLoss per batch element of shape B