This PR removes the Artifacts dependency from the main project.
Instead, it adds an Artifacts.toml file to the test project for testing purposes.
Users can use this as an example of how to use Artifacts in their own projects.
The motivation is that we should not expect users to use Artifacts.
We should offer them a universal approach to reading problem files, for example,
by expecting file paths to the problem files that want to be read.
All other libraries out there (libDAI, Merlin, toulbar2) expect file paths to the relevant problem files.
This PR also uses a different Artifact.toml file than the one used previously.
The new Artifact contains problems and solutions for the 4 tasks supported by TensorInference.jl, namely, MAR, MPE, MMAP, and PR.
This brings us a step closer to improving the correctness tests for the MPE, MMAP, and PR tasks.
The changes look good! I do not know what happened to the CI, maywe run out of quota. Since the tests pass on my local machine, I will merge it directly. Thanks!
This PR removes the
Artifacts
dependency from the main project. Instead, it adds anArtifacts.toml
file to the test project for testing purposes. Users can use this as an example of how to use Artifacts in their own projects. The motivation is that we should not expect users to use Artifacts. We should offer them a universal approach to reading problem files, for example, by expecting file paths to the problem files that want to be read. All other libraries out there (libDAI, Merlin, toulbar2) expect file paths to the relevant problem files. This PR also uses a differentArtifact.toml
file than the one used previously. The new Artifact contains problems and solutions for the 4 tasks supported by TensorInference.jl, namely, MAR, MPE, MMAP, and PR. This brings us a step closer to improving the correctness tests for the MPE, MMAP, and PR tasks.