Source code for pyro.distributions.transforms.planar

import math

import torch
import torch.nn as nn
from torch.distributions import constraints
import torch.nn.functional as F

from torch.distributions import Transform
from pyro.distributions.conditional import ConditionalTransformModule
from pyro.distributions.torch_transform import TransformModule
from pyro.distributions.util import copy_docs_from


@copy_docs_from(Transform)
class ConditionedPlanarFlow(Transform):
    def __init__(self, bias=None, u=None, w=None):
        super(ConditionedPlanarFlow, self).__init__(cache_size=1)
        self.bias = bias
        self.u = u
        self.w = w
        self._cached_logDetJ = None

    # This method ensures that torch(u_hat, w) > -1, required for invertibility
    def u_hat(self, u, w):
        alpha = torch.matmul(u.unsqueeze(-2), w.unsqueeze(-1)).squeeze(-1)
        a_prime = -1 + F.softplus(alpha)
        return u + (a_prime - alpha) * w.div(w.norm(dim=-1, keepdim=True))

    def _call(self, x):
        """
        :param x: the input into the bijection
        :type x: torch.Tensor
        Invokes the bijection x => y; in the prototypical context of a TransformedDistribution `x` is a
        sample from the base distribution (or the output of a previous flow)
        """

        # x ~ (batch_size, dim_size, 1)
        # w ~ (batch_size, 1, dim_size)
        # bias ~ (batch_size, 1)
        act = torch.tanh(torch.matmul(self.w.unsqueeze(-2), x.unsqueeze(-1)).squeeze(-1) + self.bias)
        u_hat = self.u_hat(self.u, self.w)
        y = x + u_hat * act

        psi_z = (1. - act.pow(2)) * self.w
        self._cached_logDetJ = torch.log(
            torch.abs(1 + torch.matmul(psi_z.unsqueeze(-2), u_hat.unsqueeze(-1)).squeeze(-1).squeeze(-1)))

        return y

    def _inverse(self, y):
        """
        :param y: the output of the bijection
        :type y: torch.Tensor
        Inverts y => x. As noted above, this implementation is incapable of inverting arbitrary values
        `y`; rather it assumes `y` is the result of a previously computed application of the bijector
        to some `x` (which was cached on the forward call)
        """

        raise KeyError("ConditionalPlanarFlow expected to find key in intermediates cache but didn't")

    def log_abs_det_jacobian(self, x, y):
        """
        Calculates the elementwise determinant of the log jacobian
        """
        return self._cached_logDetJ


[docs]@copy_docs_from(ConditionedPlanarFlow) class PlanarFlow(ConditionedPlanarFlow, TransformModule): """ A 'planar' normalizing flow that uses the transformation :math:`\\mathbf{y} = \\mathbf{x} + \\mathbf{u}\\tanh(\\mathbf{w}^T\\mathbf{z}+b)` where :math:`\\mathbf{x}` are the inputs, :math:`\\mathbf{y}` are the outputs, and the learnable parameters are :math:`b\\in\\mathbb{R}`, :math:`\\mathbf{u}\\in\\mathbb{R}^D`, :math:`\\mathbf{w}\\in\\mathbb{R}^D` for input dimension :math:`D`. For this to be an invertible transformation, the condition :math:`\\mathbf{w}^T\\mathbf{u}>-1` is enforced. Together with `TransformedDistribution` this provides a way to create richer variational approximations. Example usage: >>> base_dist = dist.Normal(torch.zeros(10), torch.ones(10)) >>> plf = PlanarFlow(10) >>> pyro.module("my_plf", plf) # doctest: +SKIP >>> plf_dist = dist.TransformedDistribution(base_dist, [plf]) >>> plf_dist.sample() # doctest: +SKIP tensor([-0.4071, -0.5030, 0.7924, -0.2366, -0.2387, -0.1417, 0.0868, 0.1389, -0.4629, 0.0986]) The inverse of this transform does not possess an analytical solution and is left unimplemented. However, the inverse is cached when the forward operation is called during sampling, and so samples drawn using planar flow can be scored. :param input_dim: the dimension of the input (and output) variable. :type input_dim: int References: Variational Inference with Normalizing Flows [arXiv:1505.05770] Danilo Jimenez Rezende, Shakir Mohamed """ domain = constraints.real codomain = constraints.real bijective = True event_dim = 1 def __init__(self, input_dim): super(PlanarFlow, self).__init__() self.bias = nn.Parameter(torch.Tensor(input_dim,)) self.u = nn.Parameter(torch.Tensor(input_dim,)) self.w = nn.Parameter(torch.Tensor(input_dim,)) self.input_dim = input_dim self.reset_parameters()
[docs] def reset_parameters(self): stdv = 1. / math.sqrt(self.u.size(0)) self.w.data.uniform_(-stdv, stdv) self.u.data.uniform_(-stdv, stdv) self.bias.data.uniform_(-stdv, stdv)
[docs]@copy_docs_from(ConditionalTransformModule) class ConditionalPlanarFlow(ConditionalTransformModule): """ A conditional 'planar' normalizing flow that uses the transformation :math:`\\mathbf{y} = \\mathbf{x} + \\mathbf{u}\\tanh(\\mathbf{w}^T\\mathbf{z}+b)` where :math:`\\mathbf{x}` are the inputs with dimension :math:`D`, :math:`\\mathbf{y}` are the outputs, and the pseudo-parameters :math:`b\\in\\mathbb{R}`, :math:`\\mathbf{u}\\in\\mathbb{R}^D`, and :math:`\\mathbf{w}\\in\\mathbb{R}^D` are the output of a function, e.g. a NN, with input :math:`z\\in\\mathbb{R}^{M}` representing the context variable to condition on. For this to be an invertible transformation, the condition :math:`\\mathbf{w}^T\\mathbf{u}>-1` is enforced. Together with `ConditionalTransformedDistribution` this provides a way to create richer variational approximations. Example usage: >>> from pyro.nn.dense_nn import DenseNN >>> input_dim = 10 >>> context_dim = 5 >>> batch_size = 3 >>> base_dist = dist.Normal(torch.zeros(input_dim), torch.ones(input_dim)) >>> hypernet = DenseNN(context_dim, [50, 50], param_dims=[1, input_dim, input_dim]) >>> plf = ConditionalPlanarFlow(hypernet) >>> z = torch.rand(batch_size, context_dim) >>> plf_dist = dist.ConditionalTransformedDistribution(base_dist, [plf]).condition(z) >>> plf_dist.sample(sample_shape=torch.Size([batch_size])) # doctest: +SKIP The inverse of this transform does not possess an analytical solution and is left unimplemented. However, the inverse is cached when the forward operation is called during sampling, and so samples drawn using planar flow can be scored. :param nn: a function inputting the context variable and outputting a triplet of real-valued parameters of dimensions :math:`(1, D, D)`. :type nn: callable References: Variational Inference with Normalizing Flows [arXiv:1505.05770] Danilo Jimenez Rezende, Shakir Mohamed """ domain = constraints.real codomain = constraints.real bijective = True event_dim = 1 def __init__(self, nn): super(ConditionalPlanarFlow, self).__init__() self.nn = nn
[docs] def condition(self, context): bias, u, w = self.nn(context) return ConditionedPlanarFlow(bias, u, w)