Difference between lizard and cool random forest
Can someone explain the difference between sampsize and classwt in the random forest R package part. Is there any relationship between the two parameters (if the problem is a classification issue)
thank
sampsize
represents the sample size that is drawn. Thus, it sampsize = 10
will mean that 10 samples have been collected. classwt
- weighing classes. This can be used for a weighted random forest. For example, you can use this to weigh more recent data rather than older data (see here ). Having said that, downsampling (which reduces the sample size) is seemingly just as efficient as class weighting (see here ), so you can but cannot use others as well.