DynPricing PPO V8 Experiments

Initial public release of the DynPricing PPO V8 experimental framework. This release corresponds to the final experimental configuration used in the Master's dissertation: "Dynamic Pricing in Retail: A Reinforcement Learning Approach" This release includes: Final PPO V8 configuration (ppo_tuned_v8.yaml) MDP environment implementation (simple_price_env_v8.py) Training and evaluation scripts Baseline comparison scripts Final experimental outputs: ppo_summary.csv comparison_ppo_baselines.csv best_model.zip config_used.yaml command.txt Reproducibility characteristics: Chronological data split (no temporal leakage) Fixed hyperparameters Explicit execution command log Versioned dataset (DOI: https://doi.org/10.5281/zenodo.18664650) Identical evaluation environment for PPO and baseline policies This version represents the frozen experimental state corresponding to dissertation validation. License: Academic research use.