The FS planner strives to be compatible with containerization technologies. At the moment, we support Docker and Singularity, but would be glad to extend this support to other containerization solutions should the need arise.
A ready-to-use singularity image can be built from the main FS repository as follows:
cd $FS_PATH
sudo singularity build fs-planner.img ./Singularity
The image can the be used both run on planning problems or as an interactive shell. To start a container with an interactive shell, you need to:
mkdir -p $HOME/tmp/workspace
singularity shell -C -B $DOWNWARD_BENCHMARKS/:/planning/downward-benchmarks:ro -B $HOME/tmp/workspace:/planning/workspace:rw fs-planner.img
This assumes that we have a directory $DOWNWARD_BENCHMARKS in the host machine,
which will be mapped into the /fs-planner/downward-benchmarks directory in the container
(in read-only mode, ro).
From within the container, we can now issue the standard planner commands:
/planning/fs-planner/run.py -w /planning/workspace -i /planning/downward-benchmarks/blocks/probBLOCKS-4-2.pddl --driver sbfws --options "successor_generation=adaptive,evaluator_t=adaptive,bfws.rs=sim"
The standard planner output files can then be found under the $HOME/tmp/workspace directory in the host machine.
A different possibility is to run the container directly to solve a certain planning problem. This can be achieved e.g. by issuing:
singularity run -C -H /path/to/homedir fs-planner.img domain.pddl problem.pddl solution.plan
where /path/to/homedir is the absolute path to a directory in the host machine
which will be mounted as the home directory within the container, and
which contains the domain.pddl and problem.pddl files.
The solution plan will then be left in that same directory once the planner has finished executing.
These instructions might be outdated and need to be revised
The fastest and recommended way to get started with the planner is by grabbing the ready-to-use Docker image and running it interactively.
In order to do so, you need to have Docker installed on your machine, and then you can pull the image with
docker pull aigupf/fsOnce the image is in your computer, you can log into a fully-functioning Docker container where the planner and all its software dependencies are already installed for you:
docker run -it aigupf/fs
python3 preprocessor/runner.py --tag test --instance $FSBENCHMARKS/benchmarks/counters-fn/instance_5.pddl --run --driver=smart