of :class:`~minimization.metric_gaussian_kl.MetricGaussianKL` requires an instance of

which computes the sampled estimated of the KL divergence, its gradient and the

:class:`~operators.energy_operators.StandardHamiltonian`, an operator to compute the negative log-likelihood of the problem in standardized coordinates

Fisher metric. The constructor of

at a given position in parameter space. Finally, the :class:`~operators.energy_operators.StandardHamiltonian` can be constructed from

:class:`~minimization.metric_gaussian_kl.MetricGaussianKL` requires an instance

the likelihood, represented by an :class:`~operators.energy_operators.EnergyOperator` instance. Several commonly used forms of the likelihoods are already provided in

of :class:`~operators.energy_operators.StandardHamiltonian`, an operator to

NIFTy, such as :class:`~operators.energy_operators.GaussianEnergy`, :class:`~operators.energy_operators.PoissonianEnergy`,

compute the negative log-likelihood of the problem in standardized coordinates

:class:`~operators.energy_operators.InverseGammaLikelihood` or :class:`~operators.energy_operators.BernoulliEnergy`, but the user

at a given position in parameter space.

is free to implement a likelihood customized to the problem at hand. The dome code `demos/getting_started_3.py` illustrates how to set up an energy functional

Finally, the :class:`~operators.energy_operators.StandardHamiltonian`

for MGVI and minimize it.

can be constructed from the likelihood, represented by an