Functional¶
Bayesian KL Loss¶
-
torchbnn.functional.
bayesian_kl_loss
(model, reduction='mean', last_layer_only=False)[source]¶ An method for calculating KL divergence of whole layers in the model.
Parameters: - model (nn.Module) – a model to be calculated for KL-divergence.
- reduction (string, optional) – Specifies the reduction to apply to the output:
'mean'
: the sum of the output will be divided by the number of elements of the output.'sum'
: the output will be summed. - last_layer_only (Bool) – True for return only the last layer’s KL divergence.