MyHeartCounts/openmhc-segrnn-fc

OpenMHC Forecasting — SegRNN

Track 3 (forecasting) reference checkpoint for the MyHeartCounts / OpenMHC wearable-health benchmark.

This checkpoint is a SegRNN model. SegRNN is an RNN-based long-horizon forecaster that uses segment-wise iterations and parallel multi-step decoding.

Trained from scratch on the MHC training split using the PyPOTS implementation.

• Checkpoint format: PyPOTS checkpoint (model.pypots) + standard_scaler_stats.json + training_config.json
• Forecasting task: 24-hour-ahead, 19 sensor channels, hourly resolution.

Model & implementation

• PyPOTS toolkit (implementation)
• PyPOTS documentation
• SegRNN (original implementation)
• Paper: SegRNN: Segment Recurrent Neural Network for Long-Term Time Series Forecasting (Lin et al., IEEE IoT-J 2025)

Usage

import openmhc
from openmhc.forecasters import SegRNNForecaster

# pip install "openmhc[pypots]"
fc = SegRNNForecaster.from_release("hf://MyHeartCounts/openmhc-segrnn-fc@v1.1")
results = openmhc.evaluate_forecasting(fc, version="full")

The same bundle also loads in the evaluation harness via model.release_dir=hf://MyHeartCounts/openmhc-segrnn-fc@v1.1. See openmhc_manifest.json for provenance (training run, base model, fine-tuning details) and architecture metadata.

Citation

If you use this checkpoint, please cite the OpenMHC benchmark and the original SegRNN work (linked above).