RasmussenLab / phamb

Downstream processing of VAMB binning for Viral Elucidation
MIT License
44 stars 8 forks source link

What are the criteria of RF model #52

Closed ChaoXianSen closed 11 months ago

ChaoXianSen commented 11 months ago

Hi ! What score, or threshold of RF model is used to classify bacteria from viruses?

thanks!

joacjo commented 11 months ago

Hi Chao

The RF model has been trained as described in the paper - by default the code do not filter on an arbritary cutoff or threshold but utilise the label-prediction of the model.

Best, Joachim

enryH commented 11 months ago

But this then probably means that 50 percent of the trees need to predict the contig as virus?