Rigorous.RigorousImplement

Rigorous.RigorousImplement#

Rigorous.RigorousImplement

Subprocess Coordinate Contract (Issue #80)#

When optimize_rigorously() exports data for the legacy subprocess (needs_export=True, e.g. anomaly-masked datasets), the following contract applies:

  1. Exported filenames carry original frame numbers from ssd.xr.jv (e.g. PREFIX_00032.dat), so the legacy loader sets start_file_no correctly.

  2. The restrict-lists written to trimming.txt are identity (0, N, N) — no re-trimming, since the data is already trimmed.

  3. The subprocess does NOT need elution_recognition, anomaly masks, or original trimming info — all preprocessing is already applied.

make_rigorous_decomposition_impl(decomposition, rgcurve, analysis_folder=None, niter=20, method='BH', frozen_components=None, frozen_param_groups=None, trimmed_ssd=None, clear_jobs=True, function_code=None, in_process=True, monitor=True, async_=True, progress='dashboard', max_trials=0, debug=False, _dry_run=False, ns_narrow_bounds=True, ns_adaptive_nsteps=False, ns_nsteps=None, solver_kwargs=None)#

Make a rigorous decomposition using a given RG curve.

Parameters:

  • decomposition (Decomposition) – The initial decomposition to refine (built on corrected data).

  • rgcurve (RgComponentCurve) – The Rg component curve to use for refinement.

  • analysis_folder (str, optional) – The folder to save analysis results.

  • niter (int, optional) –

    Iteration budget. Meaning depends on method:

    • 'BH': literal number of Basin-Hopping outer steps (default 20).

    • 'NS': multiplied by 7 000 to form max_ncalls for UltraNest (niter=20 → 140 000 likelihood evaluations).

    Default 20.

  • method (str, optional) – Optimization algorithm: 'BH' (Basin-Hopping, default) or 'NS' (Nested Sampling / UltraNest).

  • frozen_components (list of int, optional) – 0-based indices of protein components to freeze during optimization. Their EGH shape parameters, Rg, and UV scale will be held constant at the values from the initial decomposition.

  • trimmed_ssd (SecSaxsData, optional) – Trimmed but not baseline-corrected SSD. When provided, the optimizer fits against this data (with baseline as a free parameter) instead of the corrected data in decomposition.ssd.

  • clear_jobs (bool, optional) – If True (default), clear existing job folders before starting.

  • in_process (bool, optional) – If True (default), run the optimizer in this Python process instead of spawning a subprocess. The library-prepared optimizer (with the live dsets, base curves, and spectral vectors built above) is the one that runs — no re-derivation from disk, no parent/subprocess divergence. Set False to use the legacy subprocess path (required by the tkinter GUI; available as an escape hatch for notebook users who need process isolation). See molass-library/Copilot/DESIGN_split_optimizer_architecture.md.

  • monitor (bool, optional) – Controls the MplMonitor ipywidgets dashboard. When True (default), a live dashboard is shown whether the run is in-process or subprocess. When False, no dashboard is created — the run proceeds silently. Use monitor=False for batch / comparison runs (e.g. compare_optimization_paths) where the widget is not needed.

  • progress (str or None, optional) – Deprecated and ignored — use monitor=True/False instead. Kept in the signature only for backward compatibility.

  • debug (bool, optional) – If True, enable debug mode with additional output.

Returns:

The refined decomposition object.

Return type:

Decomposition