[dnguyen227]

Adds a MathOptAI package extension so a trained predictor can be embedded in a GDPModel as a disjunction instead of MOAI's MIP formulation.

  using DisjunctiveProgramming, MathOptAI, HiGHS

  # x[1] <= 1 ? 2x[1] - x[2] : x[1] + 3x[2] + 1
  tree = MathOptAI.BinaryDecisionTree{Float64, MathOptAI.Affine{Float64}}(
      1, 1.0, MathOptAI.Affine([2.0 -1.0], [0.0]), MathOptAI.Affine([1.0 3.0], [1.0]))

  model = GDPModel(HiGHS.Optimizer)
  @variable(model, -3 <= x[1:2] <= 3)
  y, formulation = MathOptAI.add_predictor(model, Disjunctive(tree), x)
  @objective(model, Max, y[1])
  optimize!(model, gdp_method = Hull())   # or BigM(), MBM(HiGHS.Optimizer), ...

for InfiniteGDP

  using InfiniteOpt

  model = InfiniteGDPModel(HiGHS.Optimizer)
  @infinite_parameter(model, t ∈ [0, 1], num_supports = 11)
  @variable(model, -3 <= x <= 3, Infinite(t))
  y, _ = MathOptAI.add_predictor(model, Disjunctive(tree), [x, t])  # y(t), Z(t)
  optimize!(model, gdp_method = Hull())

[pulsipher]

See below, the other thing is making sure all the test pass.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.73%. Comparing base (6ab70a6) to head (9c3d251).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #149   +/-   ##
=======================================
  Coverage   99.73%   99.73%           
=======================================
  Files          22       23    +1     
  Lines        2608     2659   +51     
=======================================
+ Hits         2601     2652   +51     
  Misses          7        7           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.