Skip to content

[Metadata] Compute the features of a problem from its model #376

Closed

Description

@ocots

Agents: read the guide for agents in #310 first. Specification: this description, #306 and its comments.

Sub-issue of #306, part 2 of 5. Independent; the documentation switch waits for #311.

Scope

features(:p): dimensions, cost kind (Mayer, Lagrange, Bolza), fixed or free initial and final times, kinds and counts of constraints, computed from the ocp (OptimalControl modelings) or from the keys of the JuMP model for a problem available in JuMP only. Never written by hand. Replaces collect_problem_data of docs/browser.jl once #311 is merged; CTBenchmarks can read it.

Done when

Refs #306, #310

Activity

  1. ocots commented on Nov 6, 2026

    @ocots
    MemberAuthor

    Taken by an agent; draft PR to follow.

  2. added a commit that references this issue on Nov 6, 2026
  3. ocots commented on Nov 6, 2026

    @ocots
    MemberAuthor

    Done in #392 (merged).

    • OptimalControlProblems.features(problem) and features(pb): dimensions, cost kind, fixed or free times, number of constraints of each kind, computed from the OptimalControl model with the CTModels functions; ExtensionError without OptimalControl.
    • New model equivalence check without a solve, in comparison: the scalarized and vector models have the same features (19 problems), shown failing on a broken model.
    • docs/browser.jl reads it; documentation build green.
    • For a future problem available in JuMP only, features will be added from its JuMP model when one exists.

    🤖 Generated with Claude Code

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions