hajirazin/learnfinance-models-ppo

LearnFinance PPO Model - v2026-02-13-450ba9c43404

Proximal Policy Optimization (PPO) portfolio allocation agent using dual forecasts (LSTM + PatchTST) as features.

Model Details

• Version: v2026-02-13-450ba9c43404
• Model Type: PPO (Proximal Policy Optimization) with dual forecasts
• Training Window: 2011-01-01 to 2026-02-13
• Symbols: 14 stocks

Components

• weights.pt - PPO actor-critic network weights
• scaler.pkl - PortfolioScaler for state normalization
• symbol_order.json - Ordered list of portfolio symbols

Metrics

• Policy Loss: -0.009968774500661226
• Value Loss: 50317.601953125
• Avg Episode Return: 0.27473973821563236
• Avg Episode Sharpe: 0.20072054963155306
• Eval Sharpe: 2.5078423330733717
• Eval CAGR: 0.6663448611882279
• Eval Max Drawdown: 0.0826743044336452

Usage

from brain_api.storage.ppo import PPOHuggingFaceModelStorage

storage = PPOHuggingFaceModelStorage(repo_id="hajirazin/learnfinance-models-ppo")
artifacts = storage.download_model(version="v2026-02-13-450ba9c43404")