Raspberry 4 image

We provde an Ubuntu based, compressed image that contains gBN and all its software dependencies.
The compressed (xz) size of the image is 3.0Gb with the uncompressed image be 14Gb in size.

To use, burn the image on an SD card that is at least 16Gb in size (use a utility such as Pi Imager that can burn the compressed image directly)
Then insert the SD card in a raspberry pi 4 and boot the pi.
You can login with user gbn and password gbn2021.
This has been created and tested on a Rpi 4 with 4GB memory, but should work on all models.

Note that the first boot might take a while longer as the image will automatically be expanded to recover all additional
space on the SD card. So if you write the image on a 64Gb card the root filesystem after the first boot will contain all 64 Gb.

The image is based on Ubuntu 20.10 desktop with a large number of dependencies for code C-language development.
Both R and SWI-Prolog are installed (from sources), along with their dependencies (via Ubuntu compiled packages, eg SWI-Prolog dependencies).
gBN also depends on Gobnilp,SCIP optimisation suite, graphviz, which are also installed.

To run each of them, SWI-Prolog has to first be started in a terminal with 
   % swipl
and then issue querries in SWI-Prolog for: (a) load the library, (b) load the desired example and (c) run the relevant code. 

   ?- use_module(library(gbn)).
   ?- [cancer(aml)].
   ?- aml.
An example complete interactions run in a terminal- with outputs: 

% swipl
Welcome to SWI-Prolog (threaded, 64 bits, version 8.3.16-12-g17113eae1)
SWI-Prolog comes with ABSOLUTELY NO WARRANTY. This is free software.
Please run ?- license. for legal details.

For online help and background, visit https://www.swi-prolog.org
For built-in help, use ?- help(Topic). or ?- apropos(Word).

?- use_module(library(gbn)).
% Loading installed R library: RColorBrewer
% Loading installed R library: cowplot
% Loading installed R library: NMF
% Loading installed R library: ggplot2
% Loading installed R library: ggpubr
% Loading installed R library: gridExtra
true.

?- [cancer(aml)].
true.

?- aml.
% Starting: aml
% Starting: fisher_nets
% Starting: fam_hmaps
% Starting: gates_nets
% Starting: svg_legend
% Finished: aml
true.

?- ls( 'aml_min60-21.01.24').
% aml_min60.bn                            aml_min60_fisher.dot
% aml_min60.dat                           aml_min60_fisher_leg.svg
% aml_min60.dot                           aml_min60_fisher.svg
% aml_min60_fams/                         aml_min60_gates_best.csv
% aml_min60_fclr.csv                      aml_min60_multi_prns.png
% aml_min60_fclr.dot                      aml_min60_prns/
% aml_min60_fclr_gated.dot                aml_min60_scnti.txt
% aml_min60_fclr_gated_leg.svg            aml_min60.set
% aml_min60_fclr_gated.svg                aml_min60_std_output.txt
% aml_min60_fclr_leg.svg                  aml_min60.svg
% aml_min60_fclr.svg                      
true.

?- halt.
There were 50 or more warnings (use warnings() to see the first 50)
%

Example screenshots

There are 3 images showing the OS image in action:

Image script

The script to create the whole image is also available: gbn_box.sh

Author

Nicos Angelopoulos, 2021/01/24