biomodhub / biomod2

BIOMOD is a computer platform for ensemble forecasting of species distributions, enabling the treatment of a range of methodological uncertainties in models and the examination of species-environment relationships.
77 stars 21 forks source link

Question about Response Curves - hide model names #449

Closed JurSeuren closed 2 months ago

JurSeuren commented 2 months ago

Hello, I am currently using biomod2 to build species distribution models for a number of reptile species.

Now that I have constructed the individual models I am curious to see the response curves of my models. My question is probably a simple one. I have successfully run the bm_PlotResponseCurves command however, and this might sound stupid, I cannot see the curves itself, only the names of the constructed models. This screenshot will probably help to see what I mean: screenshot The actual plot is somewhere above these names.

My question is a very simple one: is there a way to easily get rid of the names of all the individual models? Preferably already in the bm_PlotResponseCurves command. I suppose grouping them per pseudo-absence dataset or algorithm could also be an opportunity as this would decrease the number of lines in the plot? Thanks in advance for any suggestions!

HeleneBlt commented 2 months ago

Hello there,

It's indeed a problem ! Inside bm_PlotResponseCurves, you can play with the argument models.chosen where you can choose to show fewer models with get_built_models. For instance, as you suggest, you can separate them by sets of PA or by algorithm.

Or you can modify the theme of the plot. For example :

myResponseCurves <- bm_PlotResponseCurves(bm.out = myBiomodModelOut, 
                      models.chosen = 'all',
                      fixed.var = 'median')

myResponseCurves$plot + theme(legend.text = element_text(size=3))

You can add all the different arguments of theme like legend.spacing, legend.key.size ,.... Or even remove the legend with legend.position = 'none'.

Hope it helps, Hélène

JurSeuren commented 2 months ago

Thanks very much, Hélène!