Closed neverfox closed 4 years ago
Perhaps I'm missing something but it appears the distances of the numeric variables are not accumulating and only the distance of the last column is stored in d1 and d2 before moving on:
d1
d2
https://github.com/paobranco/UBL/blob/ce3789c7c1a8e8695519e20e3f00d9f2d2fa0a54/R/smoteRegress.R#L265
Shouldn't that be d1 <- d1 + abs(T[i, x] - newM[(i - 1) * nexs + n, x])/ranges[x]? Likewise for d2?
d1 <- d1 + abs(T[i, x] - newM[(i - 1) * nexs + n, x])/ranges[x]
Duplicate of #9
neverfox
commented
4 years ago
Perhaps I'm missing something but it appears the distances of the numeric variables are not accumulating and only the distance of the last column is stored in
d1andd2before moving on:https://github.com/paobranco/UBL/blob/ce3789c7c1a8e8695519e20e3f00d9f2d2fa0a54/R/smoteRegress.R#L265
Shouldn't that be
d1 <- d1 + abs(T[i, x] - newM[(i - 1) * nexs + n, x])/ranges[x]? Likewise for d2?