File size: 1,855 Bytes
e79560c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 | ---
license: cc-by-4.0
tags:
- rf
- microwave
- s-parameters
- fem
- palace-fem
pretty_name: rfsurrogates samples
---
<!-- generated by tools/make_model_card.py from data_index/*/manifest.jsonl; do not edit by hand -->
# rfsurrogates samples
## Mission
Give RF engineers open simulation data, tested surrogate models and the know-how to build their own.
## Status
- All records so far come from one contributor.
- Records are simulated. Nothing has been measured.
- Measured validation is a plan with no date.
- Data contributions start with a small pilot of invited people.
- How to contribute: [CONTRIBUTING.md](https://github.com/benjhardy/rfsurrogates/blob/main/CONTRIBUTING.md).
## Records
Full-wave simulation records of passive microwave devices, one folder per device family, 281 records in total.
Each record is one solved design: the geometry vector p, the complex S-parameters on the family's frequency grid,
the solver, its version, the fidelity tag, mesh statistics, compute cost, contributor and licence. The schema is
`docs/DATA_CONTRACT.md` in https://github.com/benjhardy/rfsurrogates; every record passed `contrib/check_sample.py`, and the manifest row of each
file (with its sha256) is in `data_index/<family_id>/manifest.jsonl` in that repo.
Layout: `<family_id>/palace/` training records, `<family_id>/holdout/` finer-mesh holdouts (never trained on),
`<family_id>/holdout_coarse/` the same holdout designs at the training fidelity.
### edge_coupled_bpf
Index version `8cae666b9dac`.
| split | solver | fidelity | records |
|---|---|---|---|
| holdout | palace | medium | 10 |
| train | palace | coarse | 261 |
| twin | palace | coarse | 10 |
Contributors:
- benjhardy: 261 train (coarse), 10 twin (coarse), 10 holdout (medium)
Simulated data, not measurements. Licence CC-BY-4.0: credit the contributors listed above.
|