Forum: biomod2 package is now available !
Posted by: damien georges
Content: Dear BIOMOD-users,You were eagerly waiting for it, and we are happy to say that the new version of BIOMOD called biomod2 is now online on R-Forge. Although we kept the same modelling philosophy than the former version (which we will still maintain for a while), we have made crucial changes. biomod2 is now fully object-oriented and made for running on a single species only (see vignette MultiSpeciesModelling for multi-species modelling at once). For advanced BIOMOD users, the new functions might be a bit disturbing at the beginning. Then, you will see that this new version is much more advanced and practical than the former ones. Among the novelties, the addition of MAXENT in the modelling techniques, a large range of evaluation metrics, a more refined definition of ensemble modelling and ensemble forecasting, the possibility to give presence-only data and environmental rasters to biomod2 and let it extract pseudo-absence data directly. We have created several vignettes for you to get use to this new version and a figure explaining the different ways of giving data to BIOMOD. Please bear in mind that R-Forge is a development platform, it means that this new package would experience repeated updating the next couple of weeks (correcting bugs, adding documentation, adding functionalities) so think about updating the package before each new study you will do. Last but not least, all comments are welcome! If you find a bug, if you think some documentation points are unclear, if you think about new functionalities that may be useful, just let us know ASAP. We count on you to help finalizing this new version to a very nice tool. We will then release it to CRAN by the end of July. Please remember to add your code, R-version, OS and BIOMOD-version every time you report a bug or a mistake in the vignette or help files. Hoping you will enjoy this new version of BIOMOD. With our best wishes, Damien & Wilfried |
Latest Newsbiomod2 is now devel on githubdamien georges - 2020-03-02 17:08 -
biomod2 package is now available !damien georges - 2012-07-26 13:56 -
...
|
Monitor Forum | 
Start New Thread
| RE: get_variables_importance question [ Reply ] By: damien georges on 2016-08-19 07:40 | [forum:43458] |
|
Dear Hamed, By default variable importance is not computed for ensemble models (computing time concerns).. But you can do it afterwards via smth like: ---- setwd("/tmp") rm(list = ls()) library(biomod2) ## run the ensemble modeling example code example("BIOMOD_EnsembleModeling") ## get BIOMOD_Modeling output object bm.mod <- get(load(myBiomodEM@models.out.obj@link)) ## load ensemble models em.mods.names <- BIOMOD_LoadModels(myBiomodEM) em.mods.names ## by default variable importance is not computed with ensemble models get_variables_importance(myBiomodEM) ## but you can do it a posteriori em.vi.list <- lapply(em.mods.names, function(emn) { variables_importance(get(emn), data = get_formal_data(bm.mod,'expl.var')) }) names(em.vi.list) <- em.mods.names em.vi.list ---- Hope that helps, Damien |
|
| RE: get_variables_importance question [ Reply ] By: Hamed Sangoony on 2016-08-18 16:26 | [forum:43454] |
|
Dear Damien I want to get variables importance, but for Ensembled Model. I tried this code and it didn't work: myBiomodModelEval <- getModelsEvaluations(myBiomodEM) dimnames(myBiomodModelEval) myBiomodModelEval["ROC"] getModelsVarImport(myBiomodEM) Any thoughts? Thanks Hamed |
|
| RE: get_variables_importance question [ Reply ] By: damien georges on 2016-06-21 07:10 | [forum:43303] |
|
HI Ricardo, The good news is that everything seems to work properly :) Concerning the scores we decided to report the raw vi scores in the table (not the normalized one that should sum to 1 across algorithms.. but you should do it by yourself if you wanted to report the results in a more digest way). Concerning the interpretations, I am not sure to understand exactly what you show us in the table.. It seems that you have summed up vi by species and algo.. which not make so much sense.. It would be probably better to either show a vi table variable/algo by species either a table species/variable where you have sum up the vi across algo. In any case, you should always keep the variable as a row or column of your tables. Hope that helps, Damien |
|
| RE: get_variables_importance question [ Reply ] By: Ricardo Alvarez on 2016-06-21 01:02 | [forum:43302] |
|
Hi Damien, I thought that sum of variables importance in each algorithm should be 1.0 and I get values <1 in some cases and >1 in other. However, reviewing the tutorials it seem that every variables should be sum 1.0 (not the algorithm) or not? in any case, How could compare the results between algorithm ? I'm not sure to fully understand the mean of this table. Any help will be appreciated. Cheers CTA FDA GBM GLM MAXENT canyond 0.0649 0.0287 0.1432 0.2105 0.1083 seafloor 0.1657 0.0121 0.0925 0.4666 0.0735 freshwater 0.0287 0.0048 0.0343 0.1532 0.0496 bathymetry 0.9036 0.9347 0.8064 0.8314 0.8587 |
|
| RE: get_variables_importance question [ Reply ] By: damien georges on 2016-06-20 16:39 | [forum:43299] |
|
Dear Ricardo, What do you mean by: > I've obtained values <> 1 i each algorithm Moreover, I can't have access to the screen-shot you have attached.. It seems that the file is corrupted or so.. Could you please paste in plain text the output of your VI run? Cheers, Damien |
|
| get_variables_importance question [ Reply ] By: Ricardo Alvarez on 2016-06-18 09:13 | [forum:43296] Screenshot 2016-06-18 19.09.46.png (16) downloads
|
|
Hi everyone! I've running the function get_variables_importance, because I'm interesting in analyses which are most important variables that predict distribution in my models. According with the explanantion in the package we compute a simple correlation (Pearson's by default) between references predictions and the 'shuffled' one. The return score is 1-cor(pred_ref,pred_shuffled). The highest the value, the more influence the variable has on the model. However in my analysis I've obtained values <> 1 i each algorithm. Is it function working properly? Obviously I can't compare results between my algorithm with different estimation. (i've inclued a table with the results. I'm not sure if I understand properly the explanation or something happen with my biomod2 version. Any comment will be so appreciated Cheers Ricardo |
|
