This is a tool for creating synthetic plans, the generated plans will be converted to event logs (.xes files). Basically, this is a script which wrapped up current exisiting planners and the output data are in a suitable structure and format for our goal recognition experiments.

1. For running diverse planner, CPLEX and an recommended fast-downward planner is required.

2. Required python 3 runtime environment, I recommend to build a python 3 virtual environment if you both have python 2 and python 3 installed.

3. Need to complie C++ source codes, by following instuctions (CPLEX need to be pre-installed).

4. To configure the command line environment variables in absolute path

   export DIVERSE_SCORE_COMPUTATION_PATH=/home/ubuntu/plan_generators/diversescore (VM)

   export DIVERSE_FAST_DOWNWARD_PLANNER_PATH=/home/ubuntu/plan_generators/fd-red-black-postipc2018 (VM)

   export DIVERSE_SCORE_COMPUTATION_PATH=/Users/zihangs/plan_generators/diversescore (Mac)

5. Download the dataset and put the downloaded dataset in this directory.

Before run the commands, you have to check the parameter configrations, the parameters at the top of run.py. Then just run the following commands to starts.

# activate the python 3 venv (if you don't using venv, ignore this step)
source <venv>/bin/activate

# run script
python run.py

It will take a long time to run.

Once the process is completed, check this directory, there will be a sub-directory gene_data/. All the domains, problems, tests and generated plans will be there. Then this directory will be used for our next steps for mining process models.