hajirazin/learnfinance-models-patchtst

LearnFinance PatchTST Model - v2026-08-21-0af698826abd

Close-only PatchTST transformer for predicting weekly stock returns.

Model Details

• Version: v2026-08-21-0af698826abd
• Model Type: PatchTST (Patch Time Series Transformer)
• Training Window: 2016-01-01 to 2026-08-21
• Symbols: 430 stocks

Input Channels

• Close log return (1): close_ret

Metrics

• Train Loss: 0.00047008077289721763
• Validation Loss: 0.0006476868481316729
• Baseline Loss: 0.0006328598246909678
• Best Epoch: 11
• Stopped Epoch: 19

Usage

from brain_api.storage.patchtst.huggingface import PatchTSTHalalNewHuggingFaceModelStorage
from brain_api.storage.patchtst.local import PatchTSTHalalNewModelStorage

storage = PatchTSTHalalNewHuggingFaceModelStorage(
    repo_id="hajirazin/learnfinance-models-patchtst",
    local_cache=PatchTSTHalalNewModelStorage(),
)
artifacts = storage.download_model(version="v2026-08-21-0af698826abd")