Gillespie alg 中的致命错误 -*tmp* 错误 - R

Question

我有用于 model 随机系统的 Gillespie 算法的以下变体：

gillespie_counter_two <- function(N,n,...) {
  tt = 0
  x = N$M
  S = t(N$Post - N$Pre)
  u = nrow(S)
  v = ncol(S)
  count_matrix=matrix(0,nrow=n,ncol=ncol(N$Pre))
  tvec = vector("numeric",n)
  count= vector("numeric",v)
  xmat = matrix(ncol=u,nrow=n+1)
  xmat[1,] = x
  for (i in 1:n){
    h = N$h(x,tt,...)
    if (h==0){
      k=1
      while (k <= n){
        xmat[i+k-1,]=xmat[i,]
        count_matrix[i+k-1,]=0
        tvec[i+k-1]=tvec[i-1] + (tvec[i-2]-tvec[i-3])*(k)
        k=k+1
      }
      tvec[n]= tvec[i-1] + (tvec[i-2]-tvec[i-3])*(n)
      xmat[n,]=xmat[n-1,]
      xmat[n,]=xmat[n,]
      break
    }
    else{
      tt = tt+rexp(1,sum(h))
      j = sample(v,1,prob=h)
      x = x+S[,j]
      can= vector("numeric",ncol(N$Pre))
      can[j]=1
      count=count + can
      count_matrix[i,]=count
      tvec[i] = tt
      xmat[i+1,] = x
    }
  }
  return(list(t=tvec,x=xmat,r=count_matrix))
}

当我有以下输入时

N=list()
N$M= c(x1=100,x2=1,x3=0)
N$Pre= matrix(c(1,1,0,0,1,0),nrow=2,byrow=TRUE)
N$Post=matrix(c(0,2,0,0,0,1),nrow=2,byrow=TRUE)
N$h= function(x,t,th= c(th1=0.05,th2=1)) {
  with(as.list(c(x,th)),{
    return(c(th1*x1*x2,th2*x2))
  })
}

所以我运行 Gillespie_counter_two(N,1000)，它产生以下错误

Error in `[<-`(`*tmp*`, i + k - 1, , value = 0) : subscript out of bounds
In addition: There were 50 or more warnings (use warnings() to see the first 50)

到底是怎么回事？？