|
Size: 2073
Comment: Update Vagrant instructions.
|
Size: 3008
Comment: Friendlier quickstart
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 5: | Line 5: |
| With Docker, Singularity or Vagrant you can run the latest Fast Downward release without having to compile the planner yourself. In addition, you can obtain the source code of the latest release and build it manually. The instructions below assume that your PDDL files reside in the directory `$BENCHMARKS`. | Starting with Fast Downward 19.06, we provide Docker and Singularity images as well as a recommended Vagrant configuration file in order to make the use of the planner as easy as possible. Of course, you can still build Fast Downward from the source code. We provide below a few instructions to get you started as quickly as possible, but you can find a more detailed description on [[PlannerUsage]] and [[ObtainingAndRunningFastDownward]]. |
| Line 7: | Line 7: |
| == Docker == | |
| Line 9: | Line 8: |
| Install [[https://docs.docker.com/install/|Docker]] and run | == Running the Docker image == Let us assume that you have [[https://docs.docker.com/install/|Docker]] installed on your machine and want to solve a planning problem located on the `$BENCHMARKS` directory. You can run the ''satisficing'' LAMA configuration of the planner: |
| Line 15: | Line 17: |
| This makes the local directory `$BENCHMARKS` available in the container under `/benchmarks`. Note that the `$BENCHMARKS` path needs to be absolute. | Note that this mounts the local directory `$BENCHMARKS` of your host machine under the container directory `/benchmarks`, which is the place where the containerized planner reads the problems from. The path stored in the `$BENCHMARKS` variable needs to be absolute. |
| Line 17: | Line 21: |
| == Singularity == | |
| Line 19: | Line 22: |
| Install [[https://www.sylabs.io/guides/2.6/user-guide/quick_start.html#quick-installation-steps|Singularity 2.x]] (we have not tested versions 3.x) and run | == Running the Singularity image == Let us assume that you have [[https://www.sylabs.io/guides/2.6/user-guide/quick_start.html#quick-installation-steps|Singularity 2.x]] installed on your machine (at the moment we have not tested with the newer Singularity 3.x versions), and want to solve a planning problem located on the `$BENCHMARKS` directory. You can run the same LAMA configuration as before: |
| Line 26: | Line 31: |
| The first command downloads the Singularity image into the local file `downward.simg`. The second command runs Fast Downward from this image. Unlike Docker, `$BENCHMARKS` does not have to be an absolute path. | The first command downloads the Singularity image into the local file `downward.simg`, while the second command is the one running Fast Downward from this image. Unlike Docker, `$BENCHMARKS` does not have to be an absolute path. |
| Line 28: | Line 34: |
| == Vagrant == | |
| Line 30: | Line 35: |
| * Install [[https://www.vagrantup.com/|Vagrant]]. * Download the Vagrantfile for the latest [[Releases|release]] and put it into some empty directory. * Put your PDDL files into the same directory as the Vagrantfile under the names `domain.pddl` and `problem.pddl`. * Navigate to this directory in a terminal and type `vagrant up`. |
== Using a Vagrant machine == Assuming that you have [[https://www.vagrantup.com/|Vagrant]] installed on your machine, you can use the Fast Downward Vagrantfile as follows: * Place the Vagrantfile for the desired [[Releases|release]] into some empty directory. * Put your PDDL files into the same directory as the Vagrantfile, say, under names `domain.pddl` and `problem.pddl`. * Navigate to this directory and type `vagrant up`. |
| Line 40: | Line 48: |
| * See ObtainingAndRunningFastDownward | See ObtainingAndRunningFastDownward for a complete description on how to build the planner from source. |
| Line 44: | Line 52: |
| * [[PlannerUsage|Examples]] * [[../|Documentation]] * [[ScriptUsage|Running experiments]] |
* See other ways of invoking the planner on [[PlannerUsage]]. * Read about recommended [[ScriptUsage|experiment setups]]. * Go through the full [[../|Documentation]]. |
Back to the HomePage.
Quick start
Starting with Fast Downward 19.06, we provide Docker and Singularity images as well as a recommended Vagrant configuration file in order to make the use of the planner as easy as possible. Of course, you can still build Fast Downward from the source code. We provide below a few instructions to get you started as quickly as possible, but you can find a more detailed description on PlannerUsage and ObtainingAndRunningFastDownward.
Running the Docker image
Let us assume that you have Docker installed on your machine and want to solve a planning problem located on the $BENCHMARKS directory. You can run the satisficing LAMA configuration of the planner:
1 docker run --rm -v $BENCHMARKS:/benchmarks aibasel/downward --alias lama-first /benchmarks/gripper/prob01.pddl
Note that this mounts the local directory $BENCHMARKS of your host machine under the container directory /benchmarks, which is the place where the containerized planner reads the problems from. The path stored in the $BENCHMARKS variable needs to be absolute.
Running the Singularity image
Let us assume that you have Singularity 2.x installed on your machine (at the moment we have not tested with the newer Singularity 3.x versions), and want to solve a planning problem located on the $BENCHMARKS directory. You can run the same LAMA configuration as before:
The first command downloads the Singularity image into the local file downward.simg, while the second command is the one running Fast Downward from this image. Unlike Docker, $BENCHMARKS does not have to be an absolute path.
Using a Vagrant machine
Assuming that you have Vagrant installed on your machine, you can use the Fast Downward Vagrantfile as follows:
Place the Vagrantfile for the desired release into some empty directory.
Put your PDDL files into the same directory as the Vagrantfile, say, under names domain.pddl and problem.pddl.
Navigate to this directory and type vagrant up.
Log into the Vagrant virtual machine by typing vagrant ssh.
Run downward/fast-downward.py --alias lama-first /vagrant/domain.pddl /vagrant/problem.pddl.
Type exit to leave the virtual machine, then vagrant halt to stop it.
Source code
See ObtainingAndRunningFastDownward for a complete description on how to build the planner from source.
Next steps
See other ways of invoking the planner on PlannerUsage.
Read about recommended experiment setups.
Go through the full Documentation.