Skip to content

likelihoods.rst

Latest commit

 

History

History
123 lines (75 loc) · 2.56 KB

likelihoods.rst

File metadata and controls

123 lines (75 loc) · 2.56 KB

gpytorch.likelihoods

.. automodule:: gpytorch.likelihoods

.. currentmodule:: gpytorch.likelihoods

Likelihood

.. autoclass:: Likelihood
   :special-members: __call__
   :members:

One-Dimensional Likelihoods

Likelihoods for GPs that are distributions of scalar functions. (I.e. for a specific \mathbf x we expect that f(\mathbf x) \in \mathbb{R}.)

One-dimensional likelihoods should extend :obj:`gpytorch.likelihoods._OneDimensionalLikelihood` to reduce the variance when computing approximate GP objective functions. (Variance reduction is accomplished by using 1D Gauss-Hermite quadrature rather than MC-integration).

GaussianLikelihood

.. autoclass:: GaussianLikelihood
   :members:

GaussianLikelihoodWithMissingObs

.. autoclass:: GaussianLikelihoodWithMissingObs
   :members:

FixedNoiseGaussianLikelihood

.. autoclass:: FixedNoiseGaussianLikelihood
   :members:

DirichletClassificationLikelihood

.. autoclass:: DirichletClassificationLikelihood
   :members:

BernoulliLikelihood

.. autoclass:: BernoulliLikelihood
   :members:

BetaLikelihood

.. autoclass:: BetaLikelihood
   :members:

LaplaceLikelihood

.. autoclass:: LaplaceLikelihood
   :members:

NegativeBinomialLikelihood

.. autoclass:: NegativeBinomialLikelihood
   :members:

PoissonLikelihood

.. autoclass:: PoissonLikelihood
   :members:

StudentTLikelihood

.. autoclass:: StudentTLikelihood
   :members:

OrdinalLikelihood

.. autoclass:: OrdinalLikelihood
   :members:

Multi-Dimensional Likelihoods

Likelihoods for GPs that are distributions of vector-valued functions. (I.e. for a specific \mathbf x we expect that f(\mathbf x) \in \mathbb{R}^t, where t is the number of output dimensions.)

MultitaskGaussianLikelihood

.. autoclass:: MultitaskGaussianLikelihood
   :members:

SoftmaxLikelihood

.. autoclass:: SoftmaxLikelihood
   :members: