Error message from TreeSurrogate$new

First, I used my own data set and I got "Error in model.frame.default(terms(formula, lhs = lhs, rhs = rhs, data = data, :
variable lengths differ (found for '0')" after using the TreeSurrogate function...

So I tested it again using a simple simulated data:

x1 <- rnorm(100,0,1)
x2 <- rnorm(100,0,1)
x3 <- rnorm(100,0,1)
xb <- -1 + 2*x1 + 1*x2 - 3*x3
p <- 1/(1 + exp(-xb))
y <- rbinom(n = 100, size = 1, prob = p)
y <- as.factor(y)
dt <- data.frame(x1,x2,x3,y)

rf <- randomForest(y ~ ., data = dt, ntree = 50)
dt2 <- dt[-which(names(dt) == "y")]
predictor <- Predictor$new(rf, data = dt2)
res <- TreeSurrogate$new(predictor, maxdepth = 2)

And I still got the same error message "Error in model.frame.default(terms(formula, lhs = lhs, rhs = rhs, data = data, :
variable lengths differ (found for '0')"...

No sure what I did wrong. BTW, there is no error message using the IRIS data...