paddlepaddle--paddle/python/paddle/distribution/kl.py

# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

from __future__ import annotations

import functools
import warnings
from typing import TYPE_CHECKING, TypeVar

import paddle
from paddle.distribution.bernoulli import Bernoulli
from paddle.distribution.beta import Beta
from paddle.distribution.binomial import Binomial
from paddle.distribution.categorical import Categorical
from paddle.distribution.cauchy import Cauchy
from paddle.distribution.continuous_bernoulli import ContinuousBernoulli
from paddle.distribution.dirichlet import Dirichlet
from paddle.distribution.distribution import Distribution
from paddle.distribution.exponential import Exponential
from paddle.distribution.exponential_family import ExponentialFamily
from paddle.distribution.gamma import Gamma
from paddle.distribution.geometric import Geometric
from paddle.distribution.laplace import Laplace
from paddle.distribution.lognormal import LogNormal
from paddle.distribution.multivariate_normal import MultivariateNormal
from paddle.distribution.normal import Normal
from paddle.distribution.poisson import Poisson
from paddle.distribution.uniform import Uniform
from paddle.framework import in_dynamic_mode

if TYPE_CHECKING:
    from collections.abc import Callable

    from paddle import Tensor

    _T = TypeVar('_T')

__all__ = ["register_kl", "kl_divergence"]

_REGISTER_TABLE = {}


def kl_divergence(p: Distribution, q: Distribution) -> Tensor:
    r"""
    Kullback-Leibler divergence between distribution p and q.

    .. math::

        KL(p||q) = \int p(x)log\frac{p(x)}{q(x)} \mathrm{d}x

    Args:
        p (Distribution): ``Distribution`` object. Inherits from the Distribution Base class.
        q (Distribution): ``Distribution`` object. Inherits from the Distribution Base class.

    Returns:
        Tensor, Batchwise KL-divergence between distribution p and q.

    Examples:

        .. code-block:: pycon

            >>> import paddle

            >>> p = paddle.distribution.Beta(alpha=0.5, beta=0.5)
            >>> q = paddle.distribution.Beta(alpha=0.3, beta=0.7)

            >>> print(paddle.distribution.kl_divergence(p, q))
            Tensor(shape=[], dtype=float32, place=Place(cpu), stop_gradient=True,
                0.21193528)

    """
    return _dispatch(type(p), type(q))(p, q)


def register_kl(
    cls_p: type[Distribution], cls_q: type[Distribution]
) -> Callable[[_T], _T]:
    """Decorator for register a KL divergence implementation function.

    The ``kl_divergence(p, q)`` function will search concrete implementation
    functions registered by ``register_kl``, according to multi-dispatch pattern.
    If an implementation function is found, it will return the result, otherwise,
    it will raise ``NotImplementError`` exception. Users can register
    implementation function by the decorator.

    Args:
        cls_p (type[Distribution]): The Distribution type of Instance p. Subclass derived from ``Distribution``.
        cls_q (type[Distribution]): The Distribution type of Instance q. Subclass derived from ``Distribution``.

    Examples:
        .. code-block:: pycon

            >>> import paddle

            >>> @paddle.distribution.register_kl(paddle.distribution.Beta, paddle.distribution.Beta)
            >>> def kl_beta_beta():
            ...     pass  # insert implementation here
    """
    if not issubclass(cls_p, Distribution) or not issubclass(
        cls_q, Distribution
    ):
        raise TypeError('cls_p and cls_q must be subclass of Distribution')

    def decorator(f):
        _REGISTER_TABLE[cls_p, cls_q] = f
        return f

    return decorator


def _dispatch(cls_p, cls_q):
    """Multiple dispatch into concrete implement function."""

    # find all matched super class pair of p and q
    matches = [
        (super_p, super_q)
        for super_p, super_q in _REGISTER_TABLE
        if issubclass(cls_p, super_p) and issubclass(cls_q, super_q)
    ]
    if not matches:
        raise NotImplementedError

    left_p, left_q = min(_Compare(*m) for m in matches).classes
    right_p, right_q = min(_Compare(*reversed(m)) for m in matches).classes

    if _REGISTER_TABLE[left_p, left_q] is not _REGISTER_TABLE[right_p, right_q]:
        warnings.warn(
            f'Ambiguous kl_divergence({cls_p.__name__}, {cls_q.__name__}). Please register_kl({left_p.__name__}, {right_q.__name__})',
            RuntimeWarning,
        )

    return _REGISTER_TABLE[left_p, left_q]


@functools.total_ordering
class _Compare:
    def __init__(self, *classes):
        self.classes = classes

    def __eq__(self, other):
        return self.classes == other.classes

    def __le__(self, other):
        for cls_x, cls_y in zip(self.classes, other.classes):
            if not issubclass(cls_x, cls_y):
                return False
            if cls_x is not cls_y:
                break
        return True


@register_kl(Bernoulli, Bernoulli)
def _kl_bernoulli_bernoulli(p, q):
    return p.kl_divergence(q)


@register_kl(Beta, Beta)
def _kl_beta_beta(p, q):
    return (
        (q.alpha.lgamma() + q.beta.lgamma() + (p.alpha + p.beta).lgamma())
        - (p.alpha.lgamma() + p.beta.lgamma() + (q.alpha + q.beta).lgamma())
        + ((p.alpha - q.alpha) * p.alpha.digamma())
        + ((p.beta - q.beta) * p.beta.digamma())
        + (
            ((q.alpha + q.beta) - (p.alpha + p.beta))
            * (p.alpha + p.beta).digamma()
        )
    )


@register_kl(Binomial, Binomial)
def _kl_binomial_binomial(p, q):
    return p.kl_divergence(q)


@register_kl(Dirichlet, Dirichlet)
def _kl_dirichlet_dirichlet(p, q):
    return (
        (p.concentration.sum(-1).lgamma() - q.concentration.sum(-1).lgamma())
        - ((p.concentration.lgamma() - q.concentration.lgamma()).sum(-1))
        + (
            (
                (p.concentration - q.concentration)
                * (
                    p.concentration.digamma()
                    - p.concentration.sum(-1).digamma().unsqueeze(-1)
                )
            ).sum(-1)
        )
    )


@register_kl(Categorical, Categorical)
def _kl_categorical_categorical(p, q):
    return p.kl_divergence(q)


@register_kl(Cauchy, Cauchy)
def _kl_cauchy_cauchy(p, q):
    return p.kl_divergence(q)


@register_kl(ContinuousBernoulli, ContinuousBernoulli)
def _kl_continuousbernoulli_continuousbernoulli(p, q):
    return p.kl_divergence(q)


@register_kl(Normal, Normal)
def _kl_normal_normal(p, q):
    return p.kl_divergence(q)


@register_kl(MultivariateNormal, MultivariateNormal)
def _kl_mvn_mvn(p, q):
    return p.kl_divergence(q)


@register_kl(Uniform, Uniform)
def _kl_uniform_uniform(p, q):
    return p.kl_divergence(q)


@register_kl(Laplace, Laplace)
def _kl_laplace_laplace(p, q):
    return p.kl_divergence(q)


@register_kl(Geometric, Geometric)
def _kl_geometric_geometric(p, q):
    return p.kl_divergence(q)


@register_kl(ExponentialFamily, ExponentialFamily)
def _kl_expfamily_expfamily(p, q):
    """Compute kl-divergence using `Bregman divergences <https://www.lix.polytechnique.fr/~nielsen/EntropyEF-ICIP2010.pdf>`_"""
    if not type(p) == type(q):
        raise NotImplementedError

    p_natural_params = []
    for param in p._natural_parameters:
        param = param.detach()
        param.stop_gradient = False
        p_natural_params.append(param)

    q_natural_params = q._natural_parameters

    p_log_norm = p._log_normalizer(*p_natural_params)

    try:
        if in_dynamic_mode():
            p_grads = paddle.grad(
                p_log_norm, p_natural_params, create_graph=True
            )
        else:
            p_grads = paddle.static.gradients(p_log_norm, p_natural_params)
    except RuntimeError as e:
        raise TypeError(
            "Can't compute kl_divergence({cls_p}, {cls_q}) use bregman divergence. Please register_kl({cls_p}, {cls_q}).".format(
                cls_p=type(p).__name__, cls_q=type(q).__name__
            )
        ) from e

    kl = q._log_normalizer(*q_natural_params) - p_log_norm
    for p_param, q_param, p_grad in zip(
        p_natural_params, q_natural_params, p_grads
    ):
        term = (q_param - p_param) * p_grad
        kl -= _sum_rightmost(term, len(q.event_shape))

    return kl


@register_kl(Exponential, Exponential)
def _kl_exponential_exponential(p, q):
    return p.kl_divergence(q)


@register_kl(Gamma, Gamma)
def _kl_gamma_gamma(p, q):
    return p.kl_divergence(q)


@register_kl(LogNormal, LogNormal)
def _kl_lognormal_lognormal(p, q):
    return p._base.kl_divergence(q._base)


@register_kl(Poisson, Poisson)
def _kl_poisson_poisson(p, q):
    return p.kl_divergence(q)


def _sum_rightmost(value, n):
    return value.sum(list(range(-n, 0))) if n > 0 else value