Skip to content

ecg-spiht-4x

Overview

  • Modality: ECG
  • Structure: codec
  • Compression ratio: 4×
  • Sample rate: 256 Hz
  • Config: — (operating point is fully declared in the registry; no training config)
  • Run name: ecg_spiht_256hz_04x_golden
  • HuggingFace: Ambiq/compressionkit-ecg-spiht-4x-v1.0

Dataset & License

  • Dataset: PTB-XL (dataset_id: ptb-xl)
  • License: CC BY 4.0 (open)
  • Notes: Auto-downloaded on first use.

The lifecycle runner pre-flights dataset availability before training (see #26 and the dataset contract).

Reproduction

# Single command, end-to-end.
uv run compressionkit golden run ecg-spiht-4x

# Publish the deploy package to HuggingFace (requires HF_TOKEN).
uv run compressionkit golden run ecg-spiht-4x --publish

Results land under results/ecg_spiht_256hz_04x_golden/; deploy artifacts under results/ecg_spiht_256hz_04x_golden/deploy/.

Evaluation Metrics

See the modality model zoo for the full metrics table:

Each run writes quality_scorecard.json and summary.json under its results/<run>/.

Deploy Artifacts

Every successful run produces the canonical edge deploy package:

  • spiht_config.json / spiht_app_config.h — codec parameters (language-neutral + C header).
  • c_sources/spiht.[ch] — portable C99 SPIHT reference.
  • sample_stimulus.npz / reference_vectors.npz — license-safe test frames and known-good encode/decode vectors.
  • model_card.json, deploy_manifest.json — metadata.

Customization Notes

  • This operating point is declared directly in compressionkit/experiments/registry.py (no training YAML) — add a new registry entry to explore a neighbouring operating point.
  • To resume publishing without retraining, pass --skip-train to compressionkit golden run ecg-spiht-4x --publish.