lars
bacc8763d0
CI / Lint (ruff check) (push) Successful in 30s
CI / Format (ruff format) (push) Successful in 30s
CI / Sync project version with tag (push) Has been skipped
CI / Lint (ruff check) (pull_request) Successful in 33s
CI / Type check (ty) (push) Successful in 36s
CI / Format (ruff format) (pull_request) Successful in 39s
CI / Sync project version with tag (pull_request) Has been skipped
CI / Type check (ty) (pull_request) Successful in 39s
CI / Tests (pull_request) Successful in 3m36s
CI / Tests (push) Successful in 3m51s
decode_secondaries inverted a secondary's raw predicted log_mass with inv_log_transform (exp(y) - eps) unclipped. log_mass is a raw regression output, not itself the result of log_transform, so it isn't guaranteed to land in the range that round-trips cleanly: too negative and exp(y) undershoots eps, making the result go slightly negative; too positive and exp(y) overflows float32 to inf. Either one crashes the next rollout step, since a track descended from that secondary feeds its mass back in as conditioning, and log_transform raises on a non-finite input. Clip log_mass to [log(_EPS), _LOG_MASS_MAX] before inverting, guaranteeing a finite, non-negative mass. _LOG_MASS_MAX=80.0 matches the value from the stale fix/rollout-negative-secondary-mass branch (comfortably below float32's ~88.7 overflow point, far beyond any physical particle mass a converged model would predict) — that branch had already implemented this fix but forked before gitea #35/#36 and couldn't be merged as-is, so this reimplements it fresh against current master and leaves the stale branch untouched. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>