我试图计算到一个点的最近距离。因此我创建了两个 df
创建列表 1
long <- c(52.070498, 52.370216, 52.632381, 53.201233, 51.813298, 51.92442, 52.266075, 53.219383, 52.960561, 52.367027, 51.571915, 52.516775, 51.441642, 52.221537, 52.090737, 51.985103, 51.560596)
lat <- c(4.3007, 4.895168,4.753375, 5.799913, 4.690093, 4.477733, 6.155217, 6.566502, 5.920522, 6.668492, 4.768323, 6.083022, 5.469722, 6.893662, 5.12142, 5.89873, 5.091914)
stad <- c("Den Haag", "Amsterdam", "Alkmaar", "Leeuwarden", "Dordrecht", "Rotterdam", "Deventer", "Groningen", "Heerenveen", "Almelo", "Breda", "Zwolle", "Eindhoven", "Enschede", "Utrecht", "Arnhem", "Tilburg")
list1 <- data.frame(longitude=long, latitude=lat, stad)
创建列表 2
long <- c("51.476364", "52.12379", "52.456954", "51.543197", "52.307687", "53.250184", "52.640436", "53.397875", "52.491691", "52.109272", …我有一个如下所示的数据集:
TEAM1 TEAM2 EXPG1 EXPG2 Gewonnen
ADO Den Haag Groningen 1.5950 1.2672 1
我现在尝试Gewonnen根据EXPG1和预测列EXPG2.因此,我创建了一个训练和测试集,并创建了以下模型(全部使用rcaret):
modFit <- train(Gewonnen~ EXPG1 + EXPG2, data=training, method="rf", prox=TRUE)
我现在无法制作混淆矩阵,因为我的数据有更多的参考.这是真的,因为当我这样做时:
pred <- predict(modFit, testing)
head(print)
它说: 0.5324000 0.7237333 0.2811333 0.8231000 0.8299333 0.9792000
因为我想制作一个混淆矩阵,我不能把它们变成0/1,但我觉得应该有一个选项来在模型中做到这一点.
关于我应该在这个模型中改变什么以创建0/1值的任何想法.我在文档中找不到它:
modFit <- train(Gewonnen~ EXPG1 + EXPG2, data=training, method="rf", prox=TRUE)