Forum: help

RE: Visualizing model evaluation graph considering one metric only [ Reply ]
By: Maya Guéguen on 2022-03-08 14:48

Hi Luis,

The models_scores_graph function is made to plot one metric compared to one other. It does not work with only one metric.

In the new biomod2 version (see below), the corresponding function is : bm_PlotEvalMean()
But you can use in your case the bm_PlotEvalBoxplot() function to visualize all the different metrics.

A new version of biomod2 (4.0) is available on github.
You can install it with the following commands :
library(devtools)
install_github("biomodhub/biomod2")

and check the new documentation here :
https://biomodhub.github.io/biomod2/index.html

Help messages should also be reported onto the github in order to gather everything in the same place :)
https://github.com/biomodhub/biomod2/issues

Cheers,
Maya

Visualizing model evaluation graph considering one metric only [ Reply ]
By: Luís Santiago on 2022-03-07 12:01

[forum:49360]

Hi,

I have run some models using BIOMOD_Modeling(). Here, I used TSS and ROC in the models.eval.meth argument: models.eval.meth = c("TSS", "ROC")

Then, I used models_scores_graph() to see how each type of model performed. If I do metrics = c("TSS", "ROC"), the graphs come out alright. However, when I type one metric only, say metrics = "ROC" I get the following error message: "Error in h(simpleError(msg, call)) :
error in evaluating the argument 'x' in selecting a method for function 'merge': arguments imply differing number of rows: 0, 1"

Is it a straightforward alteration I need to conduct in the function script to get the graph concerning one metric only, or do I need to run the models all over again?

Thanks in advance,
Luis