Skip to main content
. 2024 Dec 2;25(23):12942. doi: 10.3390/ijms252312942
Algorithm 1 Joint Training Procedure for DeepEnhancerPPO
  •   1:

    for epoch=1 to epochs do

  •   2:

          for each batch in data_loader do

  •   3:

                Perform forward pass to extract features

  •   4:

                current_state ← model.get_feature_vector()

  •   5:

                action ← ppo.predict(current_state)

  •   6:

                model.apply_feature_mask(action)

  •   7:

                predictions ← model(batch)

  •   8:

                Optimize feature extraction and classification modules

  •   9:

          end for

  • 10:

          Train PPO agent using experiences from the current epoch

  • 11:

    end for