With this merge requests we are now able to create a docker image for HabitatSampler.
In the docker directory there is a script to build the image and another one to start a docker container. The script to start a docker container allows the user to launch either a bash console to start an R instance or a jupyter-lab to run notebooks.
To be able to run HabitatSampler with less resources as possible we now delete the main objects after their last utilization and call R garbage collector gc(full = TRUE). Furthermore, extra arguments were added to the function multi_Class_Sampling:
- save_runs : an Habitat object and extra info (to restart a run) is saved (default TRUE)
- max_num_cores : maximum number of cores for parallelism (default 5)
Since we can't launch a web-browser from a docker container, the user can now decide if the images should be either plot via a browser or plotted inline in a notebook. In both cases the interactive plot is saved as a html file in disk.
- plot_on_browser : plot on the browser or inline in a notebook (default TRUE)
All documentation was improved according to the above changes.