StochasticModel# class torch_uncertainty.methods.StochasticModel(core_model, num_samples, probabilistic=False)[source]# freeze()[source]# Freeze all Bayesian submodules in the wrapped model. Return type: None sample(num_samples=1)[source]# Sample the wrapped model multiple times. Parameters: num_samples (int) – Number of samples to generate. Defaults to 1. Returns: Sampled model states. Return type: list[dict[str, Tensor]] unfreeze()[source]# Unfreeze all Bayesian submodules in the wrapped model. Return type: None
