Training and evaluation data for routing and wavelength assignment using multi objective reinforcement learning

This dataset contains training and evaluation data generated from multi objective RL models for bi objective and 3 objective cases and the trained models. The training_monitor.csv provides the details for the training data in terms of the rewards and accepted and processed services where each row corresponds to an episode. The evaluation_monitor.csv contains the respective evaluation data including the rewards, processed and accepted numbr of services for each episode represented by a row. The trained models are saved as best_model.zip which can be used to reproduce evaluation results.