Skip to content

todo.md

Latest commit

 

History

History
24 lines (19 loc) · 1.28 KB

todo.md

File metadata and controls

24 lines (19 loc) · 1.28 KB
Raw

Open work

What's still on the books. See raven_migration.md for the function-by- function port status (most of it is done); see IMPROVEMENTS.md for the catalogue of raven-toolbox improvements that should also be back-ported into MATLAB RAVEN.

Smaller items

  • known_issues.md — backlog of low-priority edge cases / robustness gaps / dead code from the full-codebase review. None affects correctness on well-formed inputs.
  • IMPROVEMENTS.md — items marked 💡 proposed are candidates to implement (and back-port).
  • matlab_raven_backports.md — the MATLAB ↔ raven-toolbox differences record: improvements to back-port to MATLAB RAVEN, and functionality MATLAB has that raven-toolbox deliberately doesn't (MetaCyc, dynamic FBA, ftINIT metabolomics scoring).

Upstream blockers (not raven-toolbox work, but worth tracking)

  • optlang.hybrid_interface.Configuration.clone() bug — blocks HiGHS at any scale (CI catches it in tests/test_init_solvers.py).
  • GLPK's MIP solve ignores configuration.timeout at genome scale — blocks GLPK on large MILPs.
  • Both documented in init_solver_benchmark.md with concrete fix suggestions.