rf1.outputs: Random Forest 1 Model Outputs

In akeyel/rf1: Random Forest Workflow 1 for DFMIP

Description

other.outputs$rf1 gives a list of 2 objects. The first is the mosquito results, the second is the human results. Each of these objects are lists of objects. The first $MODEL is the Random Forest Model object. This can be used to generate novel predictions. The second $TEMPORAL.ACCURACY is a summary of the cross-validation accuracy, with $OVERALL giving the overall accuracy. Other accuracy results include $RMSE, $SPEARMAN, $N, $ERROR.DF The third is $ SPATIAL.ACCURACY which performs the same cross-validation, except on spatial units instead of on years. This has the same structure as $TEMPORAL.ACCURACY. The fourth is $TEMPORAL.NULL, which contains two null model results for comparison ($NULL.MEAN and $NULL.CI) each with the same structure as $TEMPORAL.ACCURACY. The fifth is $SPATIAL.NULL, which contains null model results for spatial structure, following the same structure as $TEMPORAL.NULL. The sixth object is $RETAINED.VARS, which includes a list of independent variables that were retained in the final model. (NOTE: These can also be extracted from the $MODEL object)

akeyel/rf1 documentation built on Dec. 28, 2020, 4:48 a.m.