Update model.py

            When t = 2:

            # Here d_2 ~ p(d_2|h_1), since h1 = r_p(h_0, d_1), d_2 ~ p(d_2|d_1), we still sample d_2 based on reparameterization trick

            # It should be noted that p(d_2|d_1) is not N(0,I), due to d_mean_2 and d_logvar_0 are no longer 0,

            # It should be noted that p(d_2|d_1) is not N(0,I), due to d_mean_2 and d_logvar_2 are no longer 0,

            # but parameterized by NNs.

            # Then update h_2 by using Eq. (2), h2 = r_p(h_1, d_2),

            # So we construct the time-dependent prior of latent variable d