if path_model == MODEL_vae2_1:
state_dict = torch.load(os.path.join(path_model, 'model_state.pth'),map_location=torch.device('cpu'))
model = autoencoder
print('Loading Passed')
else:
model = WitnessComplexAutoencoder(autoencoder)
state_dict = torch.load(os.path.join(path_model, 'model_state.pth'),map_location=torch.device('cpu'))
state_dict2 = torch.load(os.path.join(os.path.join(path_norm2,exp_norm2), 'model_state.pth'))
if 'latent' not in state_dict:
state_dict['latent_norm'] = state_dict2['latent_norm'] * 0.1
print('passed')
model.load_state_dict(state_dict)
model.eval()
z = model.encode(images.float())
debugging = True
np.save('/Users/simons/PycharmProjects/MT-VAEs-TDA/src/datasets/simulated/xy_trans_l_newpers/latents_wae.np', z.detach().numpy())
# plot_2Dscatter(z.detach().numpy(), labels, path_to_save=os.path.join(path_save, 'pos_{}.pdf'.format(
# time.strftime("%Y%m%d-%H%M%S"))), title=None, show=True, palette = 'custom')
state_dict = torch.load(os.path.join(path_exp, 'model_state.pth'),
map_location=torch.device('cpu'))
state_dict2 = torch.load(
os.path.join(os.path.join(path_norm2, exp_norm2), 'model_state.pth'))
if 'latent' not in state_dict:
state_dict['latent_norm'] = state_dict2['latent_norm']*0.1
print('passed')
model.load_state_dict(state_dict)
model.eval()
dataset = MNIST_offline()
data, labels = dataset.sample(train = False)
z = model.encode(torch.Tensor(data).float())
np.save(os.path.join(path_exp,'wae_path_stdiso'),z.detach().numpy())
np.save(os.path.join(path_exp, 'labelswae_path_stdiso'), labels)
#
# df = pd.DataFrame(z)
# df['labels'] = labels
# df.to_csv(os.path.join(path_exp, '{}_latents.csv'.format('final')), index=False)
# plot_2Dscatter(z.detach().numpy(), labels, path_to_save=os.path.join(
# path_exp, '{}_latent_visualization.pdf'.format('final')), title=None, show=False,
# palette='custom2')