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: Specificity percentage for presence-only data [ Reply ] By: Lauren Fuller on 2017-03-22 09:23 | [forum:45041] |
|
Hi Damien, Thanks so much for your answer. I recently discovered the "problem" and had been meaning to post a follow-up here, sorry for not getting around to it sooner. It seems that the selection of pseudo-absence locations has a huge impact on model predictions. There is nothing technically wrong with the models produced, I have now tried with three different types of pseudo-absence data sets (one random set and two target group sets) and have different maps and response curves for each of them. This was a good lesson in the selection of pseudo-absences. I will also look into the prevalence argument and see if I can apply it, thank you. Kind regards, Lauren |
|
| RE: Specificity percentage for presence-only data [ Reply ] By: damien georges on 2017-03-22 08:19 | [forum:45040] |
|
Dear Lauren, Considering that the only thing that change between the 2 models is the PA selection it sounds fair to suspect that the issue comes from it. Either the user defined PA are mislocated either they are not well coded, ... One thing you can give a try is to force the models to equally consider presences and absences/PAs using the prevalence arg of BIOMOD_Modelling function. Cheers, Damien |
|
| RE: Specificity percentage for presence-only data [ Reply ] By: Lauren Fuller on 2017-03-01 10:13 | [forum:45038] |
|
Hi Wilfried, Thanks so much for your answer. I'm afraid I have run into a complication with my model. I am using the "user.defined" method for pseudo-absences to format the "myBiomodData" to feed into the model. I have provided the species presence coordinates, pseudoabsence coordinates and explanatory variables, which produces the "myBiomodData" object with the correct number of presences, pseudoabsences and corresponding explanatory variable values for each location. The locations of these all plot correctly. However, when I put this into the model the response curves and predicted probabilities are not right. The response curves predict zero probability of a species present at any value of the explanatory variable, and the predicted probability of a species being present is very high across most of the study area. When I use randomly selected pseudo-absences the model works fine; it produces response curves which have species predicted with different probabilities along the range of the environmental variables and predicts a more realistic distribution of species presence across the study area. The only thing which differs between these two models are the locations of the pseudo-absences, do you know why this might be causing such a discrepancy? Many thanks in advance. Lauren |
|
| RE: Specificity percentage for presence-only data [ Reply ] By: Wilfried Thuiller on 2017-01-17 12:38 | [forum:43811] |
|
Dear Lauren, Yes, this is based on the background points. Does not mean much though, but helpful at least to judge the overall quality of the models. Cheers Wilfried |
|
| Specificity percentage for presence-only data [ Reply ] By: Lauren Fuller on 2017-01-17 11:00 | [forum:43810] |
|
Hi Damien & Wilfried, I am using the MAXENT method in biomod2 to model presence-only data. I would like to report a measure of commission error for my models and have noted that the get_evaluations function reports specificity (% of absences correctly predicted). However, presence-only models cannot calculate % of absences correctly predicted, so I was wondering how this specificity measure is calculated? Is it a pseudo-specificity score using background points? Many thanks, Lauren |
|
