#include <bits/stdc++.h>

using namespace std;

int main() {
  vector<double> p(7), v{0, 1.0000000000000000, 1.0833333333333333, 1.2569444444444444, 1.5353009259259260, 1.6915991512345676, 2.0513639724794235};
  for (int i = 1; i <= 5; ++i) {
    p[i] = v[i + 1] - 1;
    for (int j = 1; j < i; ++j) p[i] -= p[j] * v[i - j + 1];
  }
  p[6] = 1 - accumulate(p.begin(), p.end() - 1, 0.);
  for (int i = 7; i <= 1000000; ++i) {
    v.emplace_back(inner_product(p.begin() + 1, p.end(), v.rbegin(), 0.) + 1);
  }
  int t;
  cin >> t;
  for (int i = 0; i < t; ++i) {
    int n;
    cin >> n;
    cout << fixed << setprecision(15) << v[n] << endl;
  }
}