import os
from typing import Optional
import numpy as np
import torch
import torch.nn.functional as F
from ...data.datasets import BaseDataset
from ..base.base_utils import ModelOutput
from ..nn import BaseDecoder, BaseEncoder
from ..normalizing_flows import IAF, IAFConfig
from ..vae import VAE
from .vae_iaf_config import VAE_IAF_Config
[docs]class VAE_IAF(VAE):
"""Variational Auto Encoder with Inverse Autoregressive Flows
(:class:`~pythae.models.normalizing_flows.IAF`).
Args:
model_config(VAE_IAF_Config): The Variational Autoencoder configuration seting the main
parameters of the model.
encoder (BaseEncoder): An instance of BaseEncoder (inheriting from `torch.nn.Module` which
plays the role of encoder. This argument allows you to use your own neural networks
architectures if desired. If None is provided, a simple Multi Layer Preception
(https://en.wikipedia.org/wiki/Multilayer_perceptron) is used. Default: None.
decoder (BaseDecoder): An instance of BaseDecoder (inheriting from `torch.nn.Module` which
plays the role of decoder. This argument allows you to use your own neural networks
architectures if desired. If None is provided, a simple Multi Layer Preception
(https://en.wikipedia.org/wiki/Multilayer_perceptron) is used. Default: None.
.. note::
For high dimensional data we advice you to provide you own network architectures. With the
provided MLP you may end up with a ``MemoryError``.
"""
def __init__(
self,
model_config: VAE_IAF_Config,
encoder: Optional[BaseEncoder] = None,
decoder: Optional[BaseDecoder] = None,
):
VAE.__init__(self, model_config=model_config, encoder=encoder, decoder=decoder)
self.model_name = "VAE_IAF"
iaf_config = IAFConfig(
input_dim=(model_config.latent_dim,),
n_made_blocks=model_config.n_made_blocks,
n_hidden_in_made=model_config.n_hidden_in_made,
hidden_size=model_config.hidden_size,
include_batch_norm=False,
)
self.iaf_flow = IAF(iaf_config)
[docs] def forward(self, inputs: BaseDataset, **kwargs):
"""
The VAE NF model
Args:
inputs (BaseDataset): The training dataset with labels
Returns:
ModelOutput: An instance of ModelOutput containing all the relevant parameters
"""
x = inputs["data"]
encoder_output = self.encoder(x)
mu, log_var = encoder_output.embedding, encoder_output.log_covariance
std = torch.exp(0.5 * log_var)
z, _ = self._sample_gauss(mu, std)
z0 = z
# Pass it through the Normalizing flows
flow_output = self.iaf_flow.inverse(z) # sampling
z = flow_output.out
log_abs_det_jac = flow_output.log_abs_det_jac
recon_x = self.decoder(z)["reconstruction"]
loss, recon_loss, kld = self.loss_function(
recon_x, x, mu, log_var, z0, z, log_abs_det_jac
)
output = ModelOutput(
recon_loss=recon_loss,
reg_loss=kld,
loss=loss,
recon_x=recon_x,
z=z,
)
return output
def loss_function(self, recon_x, x, mu, log_var, z0, zk, log_abs_det_jac):
if self.model_config.reconstruction_loss == "mse":
recon_loss = 0.5 * F.mse_loss(
recon_x.reshape(x.shape[0], -1),
x.reshape(x.shape[0], -1),
reduction="none",
).sum(dim=-1)
elif self.model_config.reconstruction_loss == "bce":
recon_loss = F.binary_cross_entropy(
recon_x.reshape(x.shape[0], -1),
x.reshape(x.shape[0], -1),
reduction="none",
).sum(dim=-1)
# starting gaussian log-density
log_prob_z0 = (
-0.5 * (log_var + torch.pow(z0 - mu, 2) / torch.exp(log_var))
).sum(dim=1)
# prior log-density
log_prob_zk = (-0.5 * torch.pow(zk, 2)).sum(dim=1)
KLD = log_prob_z0 - log_prob_zk - log_abs_det_jac
return (recon_loss + KLD).mean(dim=0), recon_loss.mean(dim=0), KLD.mean(dim=0)
def _sample_gauss(self, mu, std):
# Reparametrization trick
# Sample N(0, I)
eps = torch.randn_like(std)
return mu + eps * std, eps
[docs] def get_nll(self, data, n_samples=1, batch_size=100):
"""
Function computed the estimate negative log-likelihood of the model. It uses importance
sampling method with the approximate posterior distribution. This may take a while.
Args:
data (torch.Tensor): The input data from which the log-likelihood should be estimated.
Data must be of shape [Batch x n_channels x ...]
n_samples (int): The number of importance samples to use for estimation
batch_size (int): The batchsize to use to avoid memory issues
"""
normal = torch.distributions.MultivariateNormal(
loc=torch.zeros(self.model_config.latent_dim).to(data.device),
covariance_matrix=torch.eye(self.model_config.latent_dim).to(data.device),
)
if n_samples <= batch_size:
n_full_batch = 1
else:
n_full_batch = n_samples // batch_size
n_samples = batch_size
log_p = []
for i in range(len(data)):
x = data[i].unsqueeze(0)
log_p_x = []
for j in range(n_full_batch):
x_rep = torch.cat(batch_size * [x])
encoder_output = self.encoder(x_rep)
mu, log_var = encoder_output.embedding, encoder_output.log_covariance
std = torch.exp(0.5 * log_var)
z, eps = self._sample_gauss(mu, std)
z0 = z
# Pass it through the Normalizing flows
flow_output = self.iaf_flow.inverse(z) # sampling
z = flow_output.out
log_abs_det_jac = flow_output.log_abs_det_jac
log_q_z_given_x = (
-0.5 * (log_var + torch.pow(z0 - mu, 2) / torch.exp(log_var))
).sum(dim=1) - log_abs_det_jac
log_p_z = -0.5 * (z**2).sum(dim=-1)
recon_x = self.decoder(z)["reconstruction"]
if self.model_config.reconstruction_loss == "mse":
log_p_x_given_z = -0.5 * F.mse_loss(
recon_x.reshape(x_rep.shape[0], -1),
x_rep.reshape(x_rep.shape[0], -1),
reduction="none",
).sum(dim=-1) - torch.tensor(
[np.prod(self.input_dim) / 2 * np.log(np.pi * 2)]
).to(
data.device
) # decoding distribution is assumed unit variance N(mu, I)
elif self.model_config.reconstruction_loss == "bce":
log_p_x_given_z = -F.binary_cross_entropy(
recon_x.reshape(x_rep.shape[0], -1),
x_rep.reshape(x_rep.shape[0], -1),
reduction="none",
).sum(dim=-1)
log_p_x.append(
log_p_x_given_z + log_p_z - log_q_z_given_x
) # log(2*pi) simplifies
log_p_x = torch.cat(log_p_x)
log_p.append((torch.logsumexp(log_p_x, 0) - np.log(len(log_p_x))).item())
return np.mean(log_p)