Datasets:

---

nvidia

/

Open-H-RF

Like 78

Follow

NVIDIA 71.5k

Syncs the twente-cavitation folder on the Hub with the reviewed state on GitHub (open-h/Open-H-RF), so the card, pipeline and images match on both sides.

• README.md: the GitHub card, which adds the hero passive acoustic map and the setup schematic with italic captions, links the zea file format, fixes "This data applicable", and describes reconstruct.py as the minimum variance (Capon) passive beamformer it now is (diagonal loading 1e-2), with the example output from cavitation_bubbles_1000kPa_01mL_per_min.hdf5 and a python reconstruct.py snippet.
• pipeline.yaml: the grid and dynamic range reconstruct.py writes (200 x 300 grid, x ±10 mm, z 10–40 mm, dynamic range −20–0 dB).
• assets/: adds hero.png, Schematic_setup.png and cavitation_bubbles_1000kPa_01mL_per_min.png, and drops cavitation_bubbles_10kPa_01mL_per_min.png, which the card no longer references (the example now comes from the 1000 kPa file, as in reconstruct.py).

The card uses the Open-H-RF name and links, so this folder is left out of the rename PR (#93) and the two merge cleanly in either order. reconstruct.py isn't hosted on the Hub, so it isn't part of this PR.

GitHub PRs:

• open-h/Open-H-RF#39 — Sync datasets with tristan-deep/OpenH-RF
• open-h/Open-H-RF#55 — Rename OpenH-RF to Open-H-RF

Files changed: 6.

~ twente-cavitation/README.md
~ twente-cavitation/pipeline.yaml
+ twente-cavitation/assets/hero.png
+ twente-cavitation/assets/Schematic_setup.png
+ twente-cavitation/assets/cavitation_bubbles_1000kPa_01mL_per_min.png
- twente-cavitation/assets/cavitation_bubbles_10kPa_01mL_per_min.png