FeRx-NLME

Issues somebody may already be working on are listed last. List everything by date

• bug(estimation): run_sir / run_covariance re-read of a derived iov_occasion fit has no occasions — covariance Failed, SIR ESS 1.0, returned Ok

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 64/100

  FeRx-NLME/ferx-core#1783 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• estimation: on ctmm_pd_2state, default (auto → BOBYQA) FOCEI stops 1.5 OFV above trust_region's optimum, and S / RSR are taken off-optimum

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 55/100

  FeRx-NLME/ferx-core#1782 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• covariance: SAEM / IMP / IMPMAP report implausible SEs on non-Gaussian models (R SE 4–9, RSR up to 6e8)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 45/100

  FeRx-NLME/ferx-core#1781 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• covariance: the "S is singular" message blames "fewer subjects than free parameters" when the score is identically zero

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 76/100

  FeRx-NLME/ferx-core#1780 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• estimation: subject_eta_response_correction rebuilds the Gaussian-only Laplace cache for joint PK + TTE subjects without consulting closed_form_fixed_ebe_grad_ok

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 2/5 1-3 hours Newbie friendliness 74/100

  FeRx-NLME/ferx-core#1779 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• estimation: the FD fixed-EBE score is rough in θ for non-Gaussian models with η, so gradient optimizers stop short and report convergence

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 64/100

  FeRx-NLME/ferx-core#1778 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• api: a `?` on a String helper inside an EngineError function silently drops a ferx check code — guard the reverse leak

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 55/100

  FeRx-NLME/ferx-core#1775 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• diagnostics: give codes to the entry-point refusals that have none (TV covariate on a hazard, depot readout across a reset, CTMM, TTE/IOV simulatable, horizon, run_* inputs)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 55/100

  FeRx-NLME/ferx-core#1774 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• api: the binders (bind_from_fit, layout_from_fit, bind_theta_levels*, bind_covariate_stats) return EngineError carrying E_THETA_LEVEL_BINDING / E_COVARIATE_STATS_BINDING

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 64/100

  FeRx-NLME/ferx-core#1773 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• api: fit() / fit_from_files / run_model_* / prepare_run return EngineError, as the other entry points do since #1746

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 72/100

  FeRx-NLME/ferx-core#1772 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• fix(fitrx): a Rust-API FitOptions outside the [fit_options] domain writes a .fitrx whose sir.settings the reader rejects

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 68/100

  FeRx-NLME/ferx-core#1770 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• perf(sir): mixture SIR draws re-solve every class cold

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 45/100

  FeRx-NLME/ferx-core#1769 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• feat(fitrx): FitResult records a [mixture] fit's fitted per-class Ω/Σ overrides, so run_sir / run_covariance work after .fitrx, from ferx-r and on SAEM/IMP mixture fits

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 62/100

  FeRx-NLME/ferx-core#1765 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• api: measure θ length for fit's initial params and σ/Ω shape on predict, simulate and fit (#1615 second half)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 66/100

  FeRx-NLME/ferx-core#1764 · 1 comment · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• api: predict_survival / predict_categorical skip the covariate and level-index-column checks

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 Half a day Newbie friendliness 68/100

  FeRx-NLME/ferx-core#1763 · 1 comment · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• api: a population bound for a different same-count level binding predicts silently on the model's levels

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 52/100

  FeRx-NLME/ferx-core#1762 · 1 comment · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• epic(sir): SIR correctness, provenance and R surface — run_sir(fit) reproduces in-fit SIR, and the fit says how its intervals were made

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 5/5 Over a week Newbie friendliness 28/100

  FeRx-NLME/ferx-core#1759 · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• estimation: named target distributions for an ETA in [eta_shape] (gamma, weibull, cauchy, …) via F⁻¹(Φ(η)) (#1716 part C)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 50/100

  FeRx-NLME/ferx-core#1757 · 1 comment · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• bug(sir): the low-ESS floor probe ignores options.cancel

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 2/5 1-3 hours Newbie friendliness 84/100

  FeRx-NLME/ferx-core#1754 · 1 comment · Opened Oct 7, 2026

  Maintainers usually reply within 1 day

• refactor(parser): bind a single-endpoint [error_model] sigma by name, as per-CMT and selected already do — retire E_SIGMA_ORDER_MISMATCH [follows #1001]

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 55/100

  FeRx-NLME/ferx-core#1753 · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• refactor(estimation): run_inner_loop_warm_seeded takes the BFGS seed per call site

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 Half a day Newbie friendliness 72/100

  FeRx-NLME/ferx-core#1751 · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• perf(estimation): analytic FOCE covariance Hessian under the frozen residual variance (second-order mode response)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 38/100

  FeRx-NLME/ferx-core#1750 · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• bug(io): parameter_table (the ferx amd final-model table) prints --- for every Ω/Σ SE and positional OMEGA(i,j)/SIGMA(i) labels

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 2/5 1-3 hours Newbie friendliness 86/100

  FeRx-NLME/ferx-core#1748 · 2 comments · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• test(agents_md_canonical): catch the escaped, nested and lowercase spellings of the old guidance file

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 72/100

  FeRx-NLME/ferx-core#1737 · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• bug(npde): compute_npde_npd simulates at the declared block_sigma rho, not the fitted one

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 52/100

  FeRx-NLME/ferx-core#1733 · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• parser: pin eta_shape::PREDICTION_BLOCKS against the block registry

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 2/5 1-3 hours Newbie friendliness 74/100

  FeRx-NLME/ferx-core#1732 · 1 comment · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• parser: refuse a shape θ whose bounds leave the transform's domain (tdist ν)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 66/100

  FeRx-NLME/ferx-core#1731 · 1 comment · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• test(estimation): structural guard against raw `options.interaction` reads outside interaction_for (follow-up to #1710)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 58/100

  FeRx-NLME/ferx-core#1727 · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• bug(api): FOCE-M3 warning gates on the chain's last stage, not the last estimating stage — silent for `[foce, imp]` + imp_eval_only

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 68/100

  FeRx-NLME/ferx-core#1726 · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• parser: an ETA read twice in one product, or a negation inside a product, gets the wrong scale label

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 48/100

  FeRx-NLME/ferx-core#1724 · 1 comment · Opened Oct 6, 2026

  Maintainers usually reply within 1 day

• ferx-tools: PsN-qa-style screen for ETA shape transforms (ΔOFV per ETA) (#359 P4)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel
  enhancement

  Difficulty 5/5 Over a week Newbie friendliness 35/100

  FeRx-NLME/ferx-core#1718 · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• iov: ETA shape transforms on ETA+KAPPA — classification, kappa CV% and anchors (#359 P2)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel
  enhancement

  Difficulty 5/5 Over a week Newbie friendliness 32/100

  FeRx-NLME/ferx-core#1717 · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• saem: a typical value whose ETA is transformed is reported as having 'NO associated ETA' (#359 P0)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel
  bug

  Difficulty 4/5 3-5 days Newbie friendliness 45/100

  FeRx-NLME/ferx-core#1715 · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• level blocks: a within-subject-varying covariate that is a common factor of a funnel breaks it, so a single-column block binds an unidentified model

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 65/100

  FeRx-NLME/ferx-core#1712 · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• perf(parser): symbolic individual-parameter partials are exponential on a diamond of intermediates

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 5/5 Over a week Newbie friendliness 35/100

  FeRx-NLME/ferx-core#1699 · 1 comment · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• tooling: tools/linux-test.sh re-fetches rustup + cmake every run and shares one target volume across worktrees

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 5/5 Over a week Newbie friendliness 35/100

  FeRx-NLME/ferx-core#1694 · 1 comment · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• tooling: Ctrl-C does not stop a tools/linux-test.sh container

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 62/100

  FeRx-NLME/ferx-core#1693 · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• perf(parser): lognormal_eta_chain_exact over-declines an eta-free definition that a direct log-normal one replaces (FD where the closed form is exact)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 48/100

  FeRx-NLME/ferx-core#1691 · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• parser: root_scale does not treat an exhaustive if/else reassignment as replacing the definition (spurious Custom)

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 68/100

  FeRx-NLME/ferx-core#1689 · 1 comment · Opened Oct 5, 2026

  Maintainers usually reply within 1 day

• bug(slow-tests): per_route_lag_fit_recovers_lag fails on macOS (TVKA 5.2% off) and passes on Linux — fourth instance of #990

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 45/100

  FeRx-NLME/ferx-core#1671 · 1 comment · Opened Oct 4, 2026

  Maintainers usually reply within 1 day

• E_MISSING_BLOCK names only the first missing block

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 65/100

  FeRx-NLME/ferx-core#1664 · Opened Oct 4, 2026

  Maintainers usually reply within 1 day

• feat(api): export a per-coordinate descriptor of the packed parameter vector

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 55/100

  FeRx-NLME/ferx-core#1658 · Opened Oct 4, 2026

  Maintainers usually reply within 1 day

• predict_diag / adaptive: capped SS-equilibration notes (#867) never reported; predict() doc still points at population.warnings

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 72/100

  FeRx-NLME/ferx-core#1652 · Opened Oct 3, 2026

  Maintainers usually reply within 1 day

• A θ read only in a [mixture] logit is reported "declared but not referenced"

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 78/100

  FeRx-NLME/ferx-core#1651 · Opened Oct 3, 2026

  Maintainers usually reply within 1 day

• simulate: W_TTE_DEGENERATE_HAZARD repeats once per replicate and draw with no index, so simulate_with_uncertainty_diag can return hundreds of identical lines

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 52/100

  FeRx-NLME/ferx-core#1648 · Opened Oct 3, 2026

  Maintainers usually reply within 1 day

• covariate mu-reference `eta_shared` misses a reader reached through an earlier assignment to the group's own name — exact engine with a live data term

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 68/100

  FeRx-NLME/ferx-core#1635 · Opened Oct 3, 2026

  Maintainers usually reply within 1 day

• covariate mu-reference decline notes say a theta "stays on the numerical M-step" when a surviving group moves it

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 74/100

  FeRx-NLME/ferx-core#1634 · Opened Oct 3, 2026

  Maintainers usually reply within 1 day

• io(fitrx): the engine writes .fitrx CSV doubles at six decimal places, so Rust-written bundles are not exact

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 3/5 1-2 days Newbie friendliness 70/100

  FeRx-NLME/ferx-core#1631 · Opened Oct 2, 2026

  Maintainers usually reply within 1 day

• bug(saem): a covariate group declined because its theta anchors another η costs 10–29 OFV when that η has its own group (the pair it protects never runs) — THETA_CRCL 0.16–0.24 vs 0.556

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 4/5 3-5 days Newbie friendliness 38/100

  FeRx-NLME/ferx-core#1630 · 1 comment · Opened Oct 2, 2026

  Maintainers usually reply within 1 day

• bug(impmap): IMPMAP converges ~21 units of −2 log L above FOCEI / NONMEM IMPMAP on two_cpt_oral_cov — ω²_V1 0.176 vs 0.109, ω²_KA 0.228 vs 0.178

  Open

  ⋯Report a problem with this issue

  • Is something wrong?

  • Already taken
  • Not beginner-friendly
  • Stale or invalid
  • Something else…
  • Send Cancel

  Difficulty 5/5 Over a week Newbie friendliness 25/100

  FeRx-NLME/ferx-core#1629 · Opened Oct 2, 2026

  Maintainers usually reply within 1 day