Traktor/myenv/Lib/site-packages/torch/distributions/exp_family.py

63 lines
2.3 KiB
Python
Raw Normal View History

2024-05-26 05:12:46 +02:00
import torch
from torch.distributions.distribution import Distribution
__all__ = ["ExponentialFamily"]
class ExponentialFamily(Distribution):
r"""
ExponentialFamily is the abstract base class for probability distributions belonging to an
exponential family, whose probability mass/density function has the form is defined below
.. math::
p_{F}(x; \theta) = \exp(\langle t(x), \theta\rangle - F(\theta) + k(x))
where :math:`\theta` denotes the natural parameters, :math:`t(x)` denotes the sufficient statistic,
:math:`F(\theta)` is the log normalizer function for a given family and :math:`k(x)` is the carrier
measure.
Note:
This class is an intermediary between the `Distribution` class and distributions which belong
to an exponential family mainly to check the correctness of the `.entropy()` and analytic KL
divergence methods. We use this class to compute the entropy and KL divergence using the AD
framework and Bregman divergences (courtesy of: Frank Nielsen and Richard Nock, Entropies and
Cross-entropies of Exponential Families).
"""
@property
def _natural_params(self):
"""
Abstract method for natural parameters. Returns a tuple of Tensors based
on the distribution
"""
raise NotImplementedError
def _log_normalizer(self, *natural_params):
"""
Abstract method for log normalizer function. Returns a log normalizer based on
the distribution and input
"""
raise NotImplementedError
@property
def _mean_carrier_measure(self):
"""
Abstract method for expected carrier measure, which is required for computing
entropy.
"""
raise NotImplementedError
def entropy(self):
"""
Method to compute the entropy using Bregman divergence of the log normalizer.
"""
result = -self._mean_carrier_measure
nparams = [p.detach().requires_grad_() for p in self._natural_params]
lg_normal = self._log_normalizer(*nparams)
gradients = torch.autograd.grad(lg_normal.sum(), nparams, create_graph=True)
result += lg_normal
for np, g in zip(nparams, gradients):
result -= (np * g).reshape(self._batch_shape + (-1,)).sum(-1)
return result