oa::ml::loss::PpoLossResult

PPO loss components retained for metrics and reverse mode.

StRust public API

Declaration

pub struct PpoLossResult { pub policy_loss: Matrix, pub value_loss: Matrix, pub entropy: Matrix, pub total_loss: Matrix, }
Public Rust declaration.

Public Data Members

policy_lossMatrixfield
value_lossMatrixfield
entropyMatrixfield
total_lossMatrixfield