descriptorContributions: Descriptor contributions

Description Usage Arguments Details Examples

View source: R/rf.R

Description

Calculate average descriptor contributions to random forest models.

Usage

1

Arguments

rfModels

list containing random forest models as returned by rf()

Details

See see ?randomForest::importance for details on random forest importance metrics.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
library(dplyr)

## Retrieve file paths for example data
files <- list.files(system.file('phenotypeDataCollectionSheets',
  package = 'pdi'),full.names = TRUE)

## Prepare data
d <- map(files,readPhenotypeSheet) %>%
  map(preparePhenotypeData) %>%
  bind_rows() %>%
  siteAdjustment() %>%
   mutate(`Live crown ratio (%)` = liveCrownRatio(`Total height (m)`,
     `Lower crown height (m)`),
     `Crown condition (%)` = crownCondition(`Missing crown (%)`,
                               `Crown transparency (%)`),
     `Crown volume (m^3)` = crownVolume(`Crown radius (m)`,
                               `Total height (m)`,
                               `Lower crown height (m)`,
                               `Crown condition (%)`),
     `Bleed prevalence (%)` = bleedPrevalence(`Active bleed length (mm)`,
                               `Active bleeds`,
                               `Black staining length (mm)`,
                               `Black staining`,
                               `Diameter at breast height (m)`),
     `Agrilus exit hole density (m^-2)` = agrilusExitHoleDensity(`Agrilus exit holes`,
                               `Diameter at breast height (m)`)
)

t <- makeAnalysisTable(d)

## Generate random forest models
m <- rf(t,cls = NULL,nreps = 10)

descriptor_contributions <- m %>%
  descriptorContributions()

pdi documentation built on Aug. 19, 2020, 9:06 a.m.