I’ve read that when data is binary, the reconstruction loss is modeled by a multivariate factorized Bernoulli distribution using `torch.nn.functional.binary_cross_entropy`

, so the ELBO loss can be implemented like this:

```
def loss_function(recon_x, x, mu, logvar):
BCE = F.binary_cross_entropy(recon_x, x.view(-1, patch_size*patch_size), reduction='sum')
KLD = -0.5 * torch.sum(1 + logvar - mu.pow(2) - logvar.exp())
return BCE + KLD
```

My data is not binary, how can I implement the elbo loss correctly so that it can be converge to zero?