---
license: mit
library_name: pytorch
tags:
- sensor-time-series
- physiological-signal
- generative-model
- flow-matching
language:
- en
datasets:
- physionet/mimic-iv-ecg
---

# SensorGen

[![Paper](https://img.shields.io/badge/paper-arXiv-red)](#)
[![Webpage](https://img.shields.io/badge/website-project--page-blue)](#)
[![GitHub](https://img.shields.io/badge/code-GitHub-181717?logo=github)](https://github.com/yang-ai-lab/SensorGen)
[![HuggingFace](https://img.shields.io/badge/%F0%9F%A4%97%20HuggingFace-SensorGen-FFD21E)](https://huggingface.co/yang-ai-lab/SensorGen-SiT)
[![Python](https://img.shields.io/badge/python-3.10%2B-brightgreen)](https://www.python.org/)
[![PyTorch](https://img.shields.io/badge/PyTorch-2.4%2B-EE4C2C?logo=pytorch)](https://pytorch.org/)


This repository hosts the pre-trained checkpoints used in the SensorGen study
("*Signal or Noise? Understanding Generative Models for Real-World Sensor
Time Series*"). 

| Checkpoint | Task | Dataset |
|------------|------|---------|
| `text2ecg.pt` | Text-to-ECG | MIMIC-IV ECG |
| `bp_translation.pt` | PPG and NIBP to invasive BP | VitalDB |



## Usage

Pair this checkpoint repository with the GitHub code release at
**[yang-ai-lab/SensorGen](https://github.com/yang-ai-lab/SensorGen)**.

### Download a single checkpoint

```python
from huggingface_hub import hf_hub_download

ckpt_path = hf_hub_download(
    repo_id="yang-ai-lab/SensorGen",
    filename="text2ecg.pt",
)
```

### Download all checkpoints

```bash
hf download yang-ai-lab/SensorGen --local-dir ./ckpts
```


## Task specifications

| Task | Target `x` | C × T | `c_1` (sparse) | `c_2` (dense) |
|------|------------|-------|----------------|---------------|
| Text-to-ECG | 12-lead ECG, 10 s @ 100 Hz | 12 × 1,000 | Free-text ECG report (CLIP-encoded) | — |
| PPG → invasive BP | Arterial blood pressure, 30 s @ 50 Hz | 1 × 1,500 | 6-D non-invasive BP statistics | PPG waveform, 1 × 1,500 |


## Datasets

Neither MIMIC-IV ECG nor VitalDB are redistributed in this repository.
Credentialed access is required from the original data providers:

- **MIMIC-IV ECG** — [PhysioNet credentialed access](https://physionet.org/content/mimic-iv-ecg/)
- **VitalDB** — [vitaldb.net](https://vitaldb.net)

Preprocessing pipelines that convert the raw releases into the HDF5
layout consumed by these checkpoints are documented in the GitHub README.


## Limitations and responsible use

- The generated waveforms reflect statistical patterns in the training corpus and **must not be used for clinical
  diagnosis or as a substitute for real patient recordings**.
- These models are released for *research use*. They are not approved
  medical devices and have not been evaluated for clinical safety or
  efficacy.


## Citation

If you use any of these checkpoints, please cite the SensorGen paper:

```bibtex
@article{shuai2026sensorgen,
  title={Signal or Noise? Understanding Generative Models for Real-World Sensor Time Series},
  author={Shuai, Zitao and Xu, Zongzhe and Wu, Yuntian and Li, Sirui and Li, Tianhong and Yang, Yuzhe},
  journal={arXiv preprint arXiv:2607.04245},
  year={2026}
}
```


## License

This release is distributed under the MIT License.