I am preparing my M.Tech final-year submission on “Financial-Market Forecasting with Deep Reinforcement Learning” and need the complete project package built around precious-metal markets—specifically gold and silver. The study must demonstrate short-term price prediction and volatility prediction, trained solely on historical market data. The technical build is up to you: Python with PyTorch, TensorFlow, Stable-Baselines or a comparable stack is fine, provided the code is clean, reproducible and includes clear benchmarking against a simple baseline. Choose and justify any suitable RL algorithm (e.g., DQN, PPO, A2C), handle data ingestion and preprocessing, run back-tests on a hold-out period, and present performance metrics and visualisations (equity curve, volatility forecast plots, loss graphs). Two academic artefacts must conform to the templates I will supply: • A 60-70-page IEEE-formatted report covering literature survey, methodology, experiments, results, conclusions and references. • A 20-25-slide PowerPoint distillation for the viva. In addition to the above, deliver: - Well-commented source code and notebooks - Trained model weights - A quick-start guide that lets my laptop (standard GPU) reproduce your results without extra setup An initial walkthrough of the model’s direction during the first week will help confirm alignment with my supervisor’s expectations, after which we can finalise remaining milestones and polish the documentation.