尝试使用BNlearn创建Bayesian网络,但一直出现错误;
链接至数据集:https://drive.google.com/open?id=1VX2xkPdeHKdyYqEsD0FSm1BLu1UCtOj9eVIVfA_KJ3g
Error in check.data(data, allowed.types = discrete.data.types) : variable Variable1 must have at least two levels.
即使我的所有变量都是因素且具有多个级别,它仍然为每个变量给出了该错误。如您所见,在此示例中,我的变量“model”有4个级别。
由于我无法共享变量和数据集,因此我创建了一个小数据集以及相关代码。我遇到了相同的问题。我知道我只分享了2个变量,但我对所有变量都得到了同样的错误。
library(tidyverse)
library (bnlearn)
library(openxlsx)
DataFull <- read.xlsx("(.....)/test.xlsx", sheet = 1, startRow = 1, colNames = TRUE)
set.seed(600)
DataFull <- as_tibble(DataFull)
DataFull$Variable1 <- as.factor(DataFull$Variable1)
DataFull$TargetVar <- as.factor(DataFull$TargetVar)
DataFull <- na.omit(DataFull)
DataFull <- droplevels(DataFull)
DataFull <- DataFull[sample(nrow(DataFull)),]
Data <- DataFull[1:as.integer(nrow(DataFull)*0.70)-1,]
Datatest <- DataFull[as.integer(nrow(DataFull)*0.70):nrow(DataFull),]
nrow(Data)+nrow(Datatest)==nrow(DataFull)
FocusVar <- as.character("TargetVar")
BN.naive <- naive.bayes(Data, FocusVar)
使用str(data)
,我可以看到该变量已经有2个或更多级别:
str(Data)
Classes ‘tbl_df’, ‘tbl’ and 'data.frame': 27586 obs. of 2 variables:
$ Variable1: Factor w/ 3 levels "Small","Medium",..: 2 2 3 3 3 3 3 3 3 3 ...
$ TargetVar: Factor w/ 2 levels "Yes","No": 1 1 1 1 1 1 2 1 1 1 ...
链接至数据集:https://drive.google.com/open?id=1VX2xkPdeHKdyYqEsD0FSm1BLu1UCtOj9eVIVfA_KJ3g