terminator.utils.model.loss_fn.nlcpl_full

terminator.utils.model.loss_fn.nlcpl_full(etab, E_idx, data)

Negative log composite psuedo-likelihood Averaged nlcpl per residue, across batches p(a_i,m ; a_j,n) =

softmax [

E_s(a_i,m) + E_s(a_j,n) + E_p(a_i,m ; a_j,n) + sum_(u != m,n) [

E_p(a_i,m; A_u) + E_p(A_u, a_j,n) ]

]

Returns: averaged log likelihood per residue pair, as well as the number of edges considered