QuantEcon / jlgametheory / 31949484277

100%

push

github

ENH: Add GameTracer.jl support: `ipa_solve` and `gnm_solve` (#27)

* ENH: Add GameTracer.jl support: ipa_solve and gnm_solve

Wrap GameTracer.jl's ipa_solve and gnm_solve. Random generation of the
default perturbation ray is done on the Python side (rng argument passed
to np.random.default_rng), so that the Julia RNG state is never touched.
Additional information (ret_code and the ray used) is available via
full_output=True, returned as a NashResult.

Also bump GameTheory.jl to v0.7 (required by GameTracer.jl).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* DOC: Mention the algorithm used in the descriptions of lrsnash and hc_solve

Make the summary lines consistent with those of ipa_solve and gnm_solve,
which follow the pattern "Compute ... with the xxx algorithm".

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* TST: Make equilibrium-list comparison one-to-one

Consume matched profiles so that lists like [A, B] and [A, A] are not
treated as equal.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5 <noreply@anthropic.com>

39 of 39 new or added lines in 2 files covered. (100.0%)

75 of 75 relevant lines covered (100.0%)

1.0 hits per line