Quick start

Install

cargo install problemreductions-cli
pred --version

Rust and a native build toolchain are required. ILP solving uses the bundled HiGHS backend.

The published crate may lag behind the catalog on this site. To build the current checkout:

git clone https://github.com/CodingThrust/problem-reductions
cd problem-reductions
cargo install --path problemreductions-cli

First solve

pred create MIS --graph 0-1,1-2,2-3,3-4,4-0 -o cycle.json
pred solve cycle.json
pred evaluate cycle.json --config '[true,false,true,false,false]'

MIS is Maximum Independent Set: select as many pairwise non-adjacent vertices as possible. The graph is a cycle on five vertices. solve executes the registered ILP pipeline and maps the solution back, reporting Max(2) with a configuration such as [true, false, true, false, false]. evaluate scores a configuration of your own against the same instance. Several optimal configurations exist, so the solver’s choice may differ from yours.

Terminal session

A recording of the real CLI: discover a route, transform the instance, solve, and check the result.

Open the player · Download the cast

pred path MIS ILP --json -o paths.json
python3 -c 'import json; print(json.dumps(json.load(open("paths.json"))["paths"][0]))' > path.json
pred create MIS --graph 0-1,1-2,2-3,3-4,4-0 -o cycle.json
pred reduce cycle.json --via path.json -o reduced.json
pred solve reduced.json
pred evaluate cycle.json --config '[true,false,true,false,false]'
pred solve cycle.json

path.json contains one explicitly selected route from the returned path set. reduced.json keeps the source instance and that route, so solving the bundle recovers a source solution. The final command solves the original file through its registered ILP pipeline. Both solves and the independent evaluation return Max(2); optimal solutions may differ.