hajirazin/learnfinance-models-ppo

LearnFinance PPO Model - v2026-01-30-a616641fbcd7

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

Model Details

• Version: v2026-01-30-a616641fbcd7
• Model Type: PPO (Proximal Policy Optimization) with dual forecasts
• Training Window: 2011-01-01 to 2026-01-30
• Symbols: 15 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: -0.024572072017830494
• Value Loss: 22.851287651062012
• Avg Episode Return: 0.0
• Avg Episode Sharpe: 0.0
• Eval Sharpe: 2.200058049451263
• Eval CAGR: 0.4302113270865915
• Eval Max Drawdown: 0.08199929902522717

Usage

from brain_api.storage.ppo import PPOHuggingFaceModelStorage

storage = PPOHuggingFaceModelStorage(repo_id="hajirazin/learnfinance-models-ppo")
artifacts = storage.download_model(version="v2026-01-30-a616641fbcd7")