cci_variational_autoencoder.py [498:510]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        for i in range(n_samples):
            x1 = self.data.X_orig_test[i]
            self.model.eval().cpu()

            with torch.no_grad():
                sample_latents = []
                for param in self.data.transform_params:
                    x_transformed = transformations.transform(x1, param)
                    _, mu, log_var = self.model(x_transformed.unsqueeze(0))
                    # use mean of latent
                    z = mu
                    sample_latents.append(z)
                sample_latents = torch.cat(sample_latents)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



cci_variational_autoencoder.py [524:536]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        for i in range(n_samples):
            x1 = self.data.X_orig_test[i]
            self.model.eval().cpu()

            with torch.no_grad():
                sample_latents = []
                for param in self.data.transform_params:
                    x_transformed = transformations.transform(x1, param)
                    _, mu, log_var = self.model(x_transformed.unsqueeze(0))
                    # use mean of latent
                    z = mu
                    sample_latents.append(z)
                sample_latents = torch.cat(sample_latents)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -