Good afternoon, I need to know how to obtain non-probabilistic results in MAXENT? I need to obtain only YES and NO data (presence or absence); that is, BINARY results: 1 and 0; NOT from 0, 0.2, 0.3, 0.4, etc..... to 1 how the results MAXENT normally expresses
Thank you four your help
in R:
mybinary = my threshold
mycontinuous is the prediction derived from your model across the entire raster object and mythreshold can be obtained by functions such as optim.threshold in the now discontiued SDMTools package.
Hi,
just a quick response. You are basically asking about which treshold to use to reclassify your continous output into two classes.
MaxEnt does give you a variety of tresholds in its output, based on different theories - you just have to decide for one.
You will find good explanations online and on researchgate already, f.e.:
https://www.researchgate.net/post/Which-threshold-is-better-to-be-selected-to-output-binary-distribution-map-for-birds-in-MaxEnt-Model
Cheers
Thank you very much everyone for answering, indeed I did not find a way for the model to run and give me the results directly from Maxent in Binary, however, what I did was convert the .asci files to raster in the GIS and do a reclassification. In particular I used the threshold of 0.3 because of the distribution that I could observe from my data (result of several runs on Maxent). It is the best way and the desired binary is obtained according to the type of investigation that is being carried out. Thank you all!
Hi all, I would need to calculate TSS from maxent ountput data. Also I would like to know how can I define the threshold value.
Thanks in advance
https://www.researchgate.net/publication/337971576_Code_for_Two_alternative_evaluation_metrics_to_replace_the_true_skill_statistic_in_the_assessment_of_species_distribution_models
The code avalable at the above address provides functions to compute TSS and other metrics such as SEDI.
- Badges
- Science topic
- Similar topics
- Names