LearnFinance PatchTST Model - v2026-02-27-a2fa3e684bc6
OHLCV 5-channel PatchTST transformer model for predicting weekly stock returns.
Model Details
- Version: v2026-02-27-a2fa3e684bc6
- Model Type: PatchTST (Patch Time Series Transformer)
- Training Window: 2016-01-01 to 2026-02-27
- Symbols: 417 stocks
Input Channels (11 total)
- OHLCV log returns (5): open, high, low, close, volume
- News sentiment (1)
- Fundamentals (5): gross_margin, operating_margin, net_margin, current_ratio, debt_to_equity
Metrics
- Train Loss: 0.07482080692935326
- Validation Loss: 0.06923462471942819
- Baseline Loss: 0.07440529430807391
Usage
from brain_api.storage.huggingface import PatchTSTHuggingFaceModelStorage
storage = PatchTSTHuggingFaceModelStorage(repo_id="hajirazin/learnfinance-models-patchtst")
artifacts = storage.download_model(version="v2026-02-27-a2fa3e684bc6")
- Downloads last month
- 233
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support