Confusion about training objective

[ChengkaiYang]

Thanks for your excellent work! In the following objective, the approximating posterior of the latent variable is p_old(z|y,x1:t).The paper has said it can be calculated because the latent variable z is descrete. p_old(z | y,x1:t) = p_old(z | x1:t) * p_old(y | z,x1:t) / p_old(y,x1:t).When calculating the prior p_old(z | x1:t),the origin code wrote this:"priors = modes_pred.detach().cpu().numpy()",but when calculating p_old(y | z,x1:t),I wonder why don't add detach() operator? Thanks for your patience!