Phase 3 · roadmap
active-learning acquisition — proposes the next most-informative CGYRO run from KYRO's GP posterior (max-variance / uncertainty sampling), so expensive GPU-hours go where the surrogate is least sure. Honest: leave-one-out std↔error correlation is weak (−0.28) on the 16-point map — directional guidance that strengthens with data.
KODEX KSEEK is a fast, calibrated surrogate for the full CGYRO parameter scan — reproducing its result at inference speed, so it runs inside a real-time control loop or a design search where the full computation is far too slow to call.
One line, one contract — a prediction, its uncertainty, and whether the input is in-domain.
from kronos_ml import KSEEK
model = KSEEK() # loads the trained surrogate
y, sigma, in_domain = model.predict(x) # y = active prediction
if not in_domain: # out of its trusted region
fall_back_to_full_physics() # KSEEK abstains, never extrapolatesEvery KODEX code wraps the shared spine — KHALO for calibrated uncertainty and KGATE for the out-of-domain gate — so it reports how confident it is and abstains rather than extrapolate. Benchmarks are computed on held-out data with a fixed seed; pre-registered misses are kept, not hidden.
Part of the open kronos-ml package (Apache-2.0). Open-access deposits with citable DOIs are listed below.
Open-access deposits, each with a citable DOI: