#line 2 "template.hpp"
// #pragma GCC target("avx2")
// #pragma GCC optimize("O3")
// #pragma GCC optimize("unroll-loops")

#include <bits/stdc++.h>
using namespace std;
// https://xn--kst.jp/blog/2019/08/29/cpp-comp/
// debug methods
// usage: debug(x,y);
// vector 出力できるように修正
template <typename T>
ostream& debug_print(ostream& os, const vector<T>& v) {
    os << "[";
    for (size_t i = 0; i < v.size(); ++i) {
        os << v[i];
        if (i < v.size() - 1) os << ", ";
    }
    os << "]";
    return os;
}
template <typename T>
ostream& debug_print(ostream& os, const T& var) {
    os << var;
    return os;
}
#define CHOOSE(a) CHOOSE2 a
#define CHOOSE2(a0, a1, a2, a3, a4, x, ...) x
#define debug_1(x1) { cout << #x1 << ": "; debug_print(cout, x1) << endl; }
#define debug_2(x1, x2) { cout << #x1 << ": "; debug_print(cout, x1) << ", " << #x2 << ": "; debug_print(cout, x2) << endl; }
#define debug_3(x1, x2, x3) { cout << #x1 << ": "; debug_print(cout, x1) << ", " << #x2 << ": "; debug_print(cout, x2) << ", " << #x3 << ": "; debug_print(cout, x3) << endl; }
#define debug_4(x1, x2, x3, x4) { cout << #x1 << ": "; debug_print(cout, x1) << ", " << #x2 << ": "; debug_print(cout, x2) << ", " << #x3 << ": "; debug_print(cout, x3) << ", " << #x4 << ": "; debug_print(cout, x4) << endl; }
#define debug_5(x1, x2, x3, x4, x5) { cout << #x1 << ": "; debug_print(cout, x1) << ", " << #x2 << ": "; debug_print(cout, x2) << ", " << #x3 << ": "; debug_print(cout, x3) << ", " << #x4 << ": "; debug_print(cout, x4) << ", " << #x5 << ": "; debug_print(cout, x5) << endl; }

#ifdef LOCAL
#define debug(...) CHOOSE((__VA_ARGS__, debug_5, debug_4, debug_3, debug_2, debug_1, ~))(__VA_ARGS__)
#else
#define debug(...)
#endif

using ll = long long;
using vl = vector<ll>;
using Graph = vector<vector<ll>>;
using P = pair<ll, ll>;
#define all(v) v.begin(), v.end()
template <typename T> inline bool chmax(T &a, T b) {
    return ((a < b) ? (a = b, true) : (false));
}
template <typename T> inline bool chmin(T &a, T b) {
    return ((a > b) ? (a = b, true) : (false));
}
#define rep1(i, n) for(ll i = 1; i <= ((ll)n); ++i)
// https://trap.jp/post/1224/
template <class... T> constexpr auto min(T... a) {
    return min(initializer_list<common_type_t<T...>>{a...});
}
template <class... T> constexpr auto max(T... a) {
    return max(initializer_list<common_type_t<T...>>{a...});
}
template <class... T> void input(T &...a) { (cin >> ... >> a); }
template <class T> void input(vector<T> &a) {
    for(T &x : a)
        cin >> x;
}
void print() { cout << '\n'; }
template <class T, class... Ts> void print(const T &a, const Ts &...b) {
    cout << a;
    (cout << ... << (cout << ' ', b));
    cout << '\n';
}
void print(const string &s) {
    cout << s << '\n';
}
template <class Container, typename = void>
struct is_container : std::false_type {};
template <class Container>
struct is_container<Container, std::void_t<decltype(std::declval<Container>().begin()), decltype(std::declval<Container>().end())>> : std::true_type {};
template <class Container>
typename enable_if<is_container<Container>::value>::type print(const Container& x) {
    if (!x.empty()) {
        auto it = x.begin();
        for (; it != prev(x.end()); ++it) {
            cout << *it << " ";
        }
        cout << *it << "\n";  // 最後の要素を出力して改行
    }
}
#define INT(...)                                                               \
    int __VA_ARGS__;                                                           \
    input(__VA_ARGS__)
#define LL(...)                                                                \
    long long __VA_ARGS__;                                                     \
    input(__VA_ARGS__)
#define STR(...)                                                               \
    string __VA_ARGS__;                                                        \
    input(__VA_ARGS__)
#define REP1(a) for(ll i = 0; i < a; i++)
#define REP2(i, a) for(ll i = 0; i < a; i++)
#define REP3(i, a, b) for(ll i = a; i < b; i++)
#define REP4(i, a, b, c) for(ll i = a; i < b; i += c)
#define overload4(a, b, c, d, e, ...) e
#define rep(...) overload4(__VA_ARGS__, REP4, REP3, REP2, REP1)(__VA_ARGS__)

ll inf = 3e18;
vl dx = {1, -1, 0, 0};
vl dy = {0, 0, 1, -1};
#line 3 "math/factorial.hpp"
// https://suisen-cp.github.io/cp-library-cpp/library/math/factorial.hpp
template <class T> struct factorial {
    factorial() {};
    void ensure(const int n) {
        int sz = size(fac);
        if(sz > n) {
            return;
        }
        int new_sz = max(2 * sz, n + 1);
        fac.resize(new_sz), fac_inv.resize(new_sz);
        for(int i = sz; i < new_sz; i++) {
            if(i == 0) {
                fac[i] = 1;
                continue;
            }
            fac[i] = fac[i - 1] * i;
        }
        fac_inv[new_sz - 1] = T(1) / fac[new_sz - 1];
        for(int i = new_sz - 2; i >= sz; i--) {
            fac_inv[i] = fac_inv[i + 1] * (i + 1);
        }
        return;
    }
    T get(int i) {
        ensure(i);
        return fac[i];
    }
    T operator[](int i) { return get(i); }
    T inv(int i) {
        ensure(i);
        return fac_inv[i];
    }
    T binom(int n, int i) {
        if(n < 0 || i < 0 || n < i) {
            return T(0);
        }
        ensure(n);
        return fac[n] * fac_inv[i] * fac_inv[n - i];
    }
    T perm(int n, int i) {
        if(n < 0 || i < 0 || n < i) {
            return T(0);
        }
        ensure(n);
        return fac[n] * fac_inv[n - i];
    }

  private:
    vector<T> fac, fac_inv;
};
#line 3 "poly/shift_of_sampling_points.hpp"
#include <atcoder/convolution>
// https://suisen-cp.github.io/cp-library-cpp/library/polynomial/shift_of_sampling_points.hpp
template <class Mint>
vector<Mint> shift_of_sampling_points(const vector<Mint> &f, const int m,
                                      const int c) {
    // n次未満の多項式fのn個の点f(0),...,f(n-1)に対して
    // f(c),...,f(c+m-1)を計算
    factorial<Mint> fac;
    const int n = f.size();
    auto a = [&] {
        vector<Mint> a1(n), a2(n);
        rep(i, n) {
            a1[i] = f[i] * fac.inv(i);
            a2[i] = ((i & 1) ? -1 : 1) * fac.inv(i);
        }
        auto a = convolution(a1, a2);
        a.resize(n);
        return a;
    }();
    vector<Mint> a_(n);
    rep(i, n) { a_[i] = a[n - 1 - i] * fac[n - 1 - i]; }
    vector<Mint> b2 = [&] {
        // b2[i] = c*(c-1)*...*(c-i+1)/i!
        vector<Mint> b2(n);
        Mint ci = 1;
        b2[0] = ci;
        for(int i = 1; i < n; i++) {
            ci *= c - i + 1;
            b2[i] = ci * fac.inv(i);
        }
        return b2;
    }();
    auto b = convolution(a_, b2);
    vector<Mint> fc = [&] {
        vector<Mint> f1(n);
        vector<Mint> f2(m);
        rep(i, n) { f1[i] = b[n - 1 - i] * fac.inv(i); }
        rep(i, m) { f2[i] = fac.inv(i); }
        return convolution(f1, f2);
    }();
    vector<Mint> res(m);
    rep(i, m) { res[i] = fac[i] * fc[i]; }
    return res;
}
#line 4 "math/factorial_large.hpp"
#include <atcoder/modint>
template <class mint> struct factorial_large {
    const ll K = 9;
    vector<vector<mint>> f;
    vector<mint> g;
    factorial_large() {
        // f_i(x) = (2^i x + 1) * ... * (2^i x + 2^i - 1)
        // f_i(0) , ... , f_i(2^i - 1) の値が分かればシフトできる
        f = vector(K + 1, vector<mint>());
        f[0] = {1};
        // f_0(x) = 1
        rep(i, K) {
            // cul f_(i+1)
            ll ti = 1LL << i;
            auto f1 = shift_of_sampling_points<mint>(f[i], 3 * ti, ti);
            f[i].insert(f[i].end(), all(f1));
            f[i + 1].resize(2 * ti);
            rep(j, 2 * ti) {
                f[i + 1][j] = f[i][2 * j] * f[i][2 * j + 1] * ti * (2 * j + 1);
            }
        }
        // g_i = (i*2^K)!
        {
            ll sz = ll(mint::mod()) / (1LL << K) + 1;
            auto g1 = shift_of_sampling_points(f[K], sz - 1, 0);
            g.resize(sz);
            g[0] = 1;
            mint tK = 1 << K;
            for(int i = 1; i < sz; i++) {
                g[i] = g[i - 1] * g1[i - 1] * tK * i;
            }
        }
    }
    mint fac(ll n) {
        if(n >= mint::mod())
            return 0;
        ll r = n / (1LL << K);
        ll q = n - (1LL << K) * r;
        mint res = g[r];
        for(ll i = (1LL << K) * r + 1; i <= n; i++) {
            res *= i;
        }
        return res;
    }
};
#line 5 "/home/y_midori/cp/yukicoder/455/c.cpp"
// using namespace atcoder;
using mint = atcoder::modint998244353;
factorial<mint> fac;
factorial_large<mint> fac_large;
mint cul_fac(ll n) {
    if(n < 1e6) {
        return fac[n];
    } else
        return fac_large.fac(n);
}
void solve() {
    LL(k);
    ll sum = 0;
    mint ans = 1;
    rep(i, k) {
        LL(l, m);
        sum += l * m;
        ans *= cul_fac(m).inv();
        ans *= cul_fac(l).inv().pow(m);
    }
    ans *= cul_fac(sum);
    print(ans.val());
}
int main() {
    ios::sync_with_stdio(false);
    std::cin.tie(nullptr);
    cout << std::setprecision(16);
    int t = 1;
    rep(_, t) {
        solve();
    }
}