-
Notifications
You must be signed in to change notification settings - Fork 0
/
Fig 3b and c_Random forest analysis code.R
70 lines (50 loc) · 2.36 KB
/
Fig 3b and c_Random forest analysis code.R
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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
#Choose data file
otu <- as.data.frame(read.csv(file.choose()))
otu1 <- na.omit(otu)
library(randomForest)
library(rfPermute)
from_reg1 <- ETT_ATT ~ Area + Area.depth + Max_depth + Volume + Res_time +
Elevation + Wshd + Agriculture + Development + Water + Forest +
Wetland + Grass + Shrubland + WS + TSP + SR + LR + H + LE + SuAT +
SpAT + FaAT + WiAT + Humidity + Secchi_trend
from_reg1
set.seed(123)
otu_forest1 <- randomForest(from_reg1, data = otu1, importance = TRUE, ntree = 500)
otu_forest1
###Importance of variables in explaining the ETT-ATT explored using random forest analysis
set.seed(123)
otu_rfP1 <- rfPermute(from_reg1, data = otu1, importance = TRUE, ntree = 500, nrep = 1000, num.cores = 1)
otu_rfP1
#The mean squared error (MSE)
importance_otu.scale1 <- data.frame(importance(otu_rfP1, scale = TRUE), check.names = FALSE)
importance_otu.scale1
library(A3)
#model.fn=randomForest 调用随机森林的方法进行运算
#p.acc=0.001 表示基于 1000 次随机置换获得对 p 值的估计，p.acc 值越小代表置换次数越多，运算也就越慢，因此如果对全模型 p 值不是很迫切的话还是慎用
#model.args 用于传递参数给 randomForest()，因此里面的参数项根据 randomForest() 的参数项而定，具体可 ?randomForest
#其它详情可 ?a3 查看帮助
set.seed(123)
otu_forest.pval <- a3(from_reg1, data = otu1, model.fn = randomForest, p.acc = 0.001, model.args = list(importance = TRUE, ntree = 500))
otu_forest.pval
####计算HTT_ATT
otu <- as.data.frame(read.csv(file.choose()))
otu2 <- na.omit(otu)
from_reg2 <- HTT_ATT ~ Area + Max_depth + Area.depth + Volume + Res_time +
Elevation + Wshd + Agriculture + Development + Forest + Water +
Wetland + Grass + Shrubland + WS + TSP + SR + LR + H + LE +
SuAT + SpAT + FaAT + WiAT + Humidity + Secchi_trend
from_reg2
set.seed(123)
otu_forest2 <- randomForest(from_reg2, data = otu2, importance = TRUE, ntree = 500)
otu_forest2
###Importance of variables in explaining the HTT-ATT explored using random forest analysis
set.seed(123)
otu_rfP2 <- rfPermute(from_reg2, data = otu2, importance = TRUE, ntree = 500, nrep = 1000, num.cores = 1)
otu_rfP2
#The mean squared error (MSE)
importance_otu.scale2 <- data.frame(importance(otu_rfP2, scale = TRUE), check.names = FALSE)
importance_otu.scale2
####A3 PACKAGE
set.seed(123)
otu_forest.pval <- a3(from_reg2, data = otu2, model.fn = randomForest, p.acc = 0.001, model.args = list(importance = TRUE, ntree = 500))
otu_forest.pval