The RNAforest model as shown in the publication is trained with an aging version of XGBoost and sklearn, which is getting more and more difficult to support. After taking user feedback into consideration, it will be simpler and more quantitative to deprecate RNAforest in favor of directly interpreting quantitative spatial point features e.g. proximity, polarization, etc. metrics already built in. This will likely happen in the next few versions.
The RNAforest model as shown in the publication is trained with an aging version of XGBoost and sklearn, which is getting more and more difficult to support. After taking user feedback into consideration, it will be simpler and more quantitative to deprecate RNAforest in favor of directly interpreting quantitative spatial point features e.g. proximity, polarization, etc. metrics already built in. This will likely happen in the next few versions.