ASU ABM club adaptation of the NetLogo wolf sheep predation model with added kill tracking from @mmannin5
bash (That means old versions of Windows (< 10) definitely will not work without cygwin or creating
run.sh in a different language. Linux and Mac will work fine).
- Install docker
- Clone this repository
- Build the Docker images by running
./run.sh buildon the command line in the root of the project. Alternatively, images can be downloaded from DockerHub (
docker pull comses/resbaz-analysis:3.3.3and
docker pull comses/resbaz-netlogo:5.3.1)
Run the workflow
% ./run.sh run on the command line (caveat: only tested on Linux and Mac, it may run on Windows 10 bash) will execute the entire pipeline which currently consists of (1) a NetLogo behavior space experiment that generates a single CSV file, and (2) a RMarkdown file that loads the generated CSV file and generates a RMarkdown HTML report.
The NetLogo data can be found at
data/vary_food_gains.csv and the resulting RMarkdown report will be produced at
results/wolf_sheep_AB.html, relative to the root directory.
Interact with the workflow
- NetLogo GUI should have popped up on your desktop
- If it hasn't check the logs (
docker-compose logs -f netlogo). If there is an error about not being able to connect to the X11 server window you need to give docker permission to talk to the X11 socket. This can be done with
- RStudio GUI available at
Interact with NetLogo
The root project directory is mounted at
/code so the NetLogo model used to generate this model is at
/code/src/wolf-sheep-predation.nlogo in NetLogo to view the model. The Behaviour Space settings used the
vary_food_gains experiment. Parameters for the experiment can be seen by going into Tools > BehaviourSpace and opening
Interact with RStudio
localhost:8787 login as rstudio with a password of rstudio. The root project directory (the directory this is in) is mounted into RStudio at
/home/rstudio/code. It should be visible in the explorer pane in RStudio.
wolf-sheep.Rprojproject file in RStudio (
File > Open Project...and select
wolf-sheep.Rprojit will at the path
/home/rstudio/code/wolf-sheep.Rproj). This will load the Packrat repository used for this project.
This workflow has hard coded the output file generated from NetLogo into a path made accessible to the RMarkdown container. This example uses the hard coded data set name
vary_food_gains.csv. A more robust workflow would define this type of data in a single place and referencing it in dependent computations so that changing the filename of a dataset does not require manually editing multiple files.
Wilensky, U. (1997). NetLogo Wolf Sheep Predation model. http://ccl.northwestern.edu/netlogo/models/WolfSheepPredation. Center for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL.