#397051 (C++17) No.840 ほむほむほむら

• 問題: No.840 ほむほむほむら
• ユーザー: ei1333333
• 提出日時: 2019-11-09 02:00:06
• 言語: C++17 (gcc 12.3.0 + boost 1.83.0)
• 結果: AC
• 実行時間: 5 ms / 4,000 ms
• コード長: 19,495 bytes
• コンパイル時間: 3,824 ms
• コンパイル使用メモリ: 242,500 KB
• 実行使用メモリ: 5,376 KB
• 最終ジャッジ日時: 2024-09-15 03:49:15
• 合計ジャッジ時間: 4,627 ms
• ジャッジサーバーID （参考情報）: judge1 / judge4

入力	結果	実行時間; 実行使用メモリ
testcase_00	AC	2 ms; 5,248 KB
testcase_01	AC	2 ms; 5,376 KB
testcase_02	AC	2 ms; 5,376 KB
testcase_03	AC	3 ms; 5,376 KB
testcase_04	AC	2 ms; 5,376 KB
testcase_05	AC	2 ms; 5,376 KB
testcase_06	AC	2 ms; 5,376 KB
testcase_07	AC	2 ms; 5,376 KB
testcase_08	AC	3 ms; 5,376 KB
testcase_09	AC	2 ms; 5,376 KB
testcase_10	AC	2 ms; 5,376 KB
testcase_11	AC	2 ms; 5,376 KB
testcase_12	AC	2 ms; 5,376 KB
testcase_13	AC	4 ms; 5,376 KB
testcase_14	AC	3 ms; 5,376 KB
testcase_15	AC	2 ms; 5,376 KB
testcase_16	AC	2 ms; 5,376 KB
testcase_17	AC	3 ms; 5,376 KB
testcase_18	AC	5 ms; 5,376 KB
testcase_19	AC	5 ms; 5,376 KB
testcase_20	AC	2 ms; 5,376 KB
testcase_21	AC	2 ms; 5,376 KB
testcase_22	AC	2 ms; 5,376 KB
testcase_23	AC	5 ms; 5,376 KB
testcase_24	AC	2 ms; 5,376 KB
testcase_25	AC	2 ms; 5,376 KB
testcase_26	AC	2 ms; 5,376 KB
testcase_27	AC	5 ms; 5,376 KB

ソースコード

#include <bits/stdc++.h>

using namespace std;

using int64 = long long;
const int mod = 998244353;

const int64 infll = (1LL << 62) - 1;
const int inf = (1 << 30) - 1;

struct IoSetup {
  IoSetup() {
    cin.tie(nullptr);
    ios::sync_with_stdio(false);
    cout << fixed << setprecision(10);
    cerr << fixed << setprecision(10);
  }
} iosetup;


template< typename T1, typename T2 >
ostream &operator<<(ostream &os, const pair< T1, T2 > &p) {
  os << p.first << " " << p.second;
  return os;
}

template< typename T1, typename T2 >
istream &operator>>(istream &is, pair< T1, T2 > &p) {
  is >> p.first >> p.second;
  return is;
}

template< typename T >
ostream &operator<<(ostream &os, const vector< T > &v) {
  for(int i = 0; i < (int) v.size(); i++) {
    os << v[i] << (i + 1 != v.size() ? " " : "");
  }
  return os;
}

template< typename T >
istream &operator>>(istream &is, vector< T > &v) {
  for(T &in : v) is >> in;
  return is;
}

template< typename T1, typename T2 >
inline bool chmax(T1 &a, T2 b) { return a < b && (a = b, true); }

template< typename T1, typename T2 >
inline bool chmin(T1 &a, T2 b) { return a > b && (a = b, true); }

template< typename T = int64 >
vector< T > make_v(size_t a) {
  return vector< T >(a);
}

template< typename T, typename... Ts >
auto make_v(size_t a, Ts... ts) {
  return vector< decltype(make_v< T >(ts...)) >(a, make_v< T >(ts...));
}

template< typename T, typename V >
typename enable_if< is_class< T >::value == 0 >::type fill_v(T &t, const V &v) {
  t = v;
}

template< typename T, typename V >
typename enable_if< is_class< T >::value != 0 >::type fill_v(T &t, const V &v) {
  for(auto &e : t) fill_v(e, v);
}

template< class T >
struct Matrix {
  vector< vector< T > > A;

  Matrix() {}

  Matrix(size_t n, size_t m) : A(n, vector< T >(m, 0)) {}

  Matrix(size_t n) : A(n, vector< T >(n, 0)) {};

  size_t height() const {
    return (A.size());
  }

  size_t width() const {
    return (A[0].size());
  }

  inline const vector< T > &operator[](int k) const {
    return (A.at(k));
  }

  inline vector< T > &operator[](int k) {
    return (A.at(k));
  }

  static Matrix I(size_t n) {
    Matrix mat(n);
    for(int i = 0; i < n; i++) mat[i][i] = 1;
    return (mat);
  }

  Matrix &operator+=(const Matrix &B) {
    size_t n = height(), m = width();
    assert(n == B.height() && m == B.width());
    for(int i = 0; i < n; i++)
      for(int j = 0; j < m; j++)
        (*this)[i][j] += B[i][j];
    return (*this);
  }

  Matrix &operator-=(const Matrix &B) {
    size_t n = height(), m = width();
    assert(n == B.height() && m == B.width());
    for(int i = 0; i < n; i++)
      for(int j = 0; j < m; j++)
        (*this)[i][j] -= B[i][j];
    return (*this);
  }

  Matrix &operator*=(const Matrix &B) {
    size_t n = height(), m = B.width(), p = width();
    assert(p == B.height());
    vector< vector< T > > C(n, vector< T >(m, 0));
    for(int i = 0; i < n; i++)
      for(int j = 0; j < m; j++)
        for(int k = 0; k < p; k++)
          C[i][j] = (C[i][j] + (*this)[i][k] * B[k][j]);
    A.swap(C);
    return (*this);
  }

  Matrix &operator^=(long long k) {
    Matrix B = Matrix::I(height());
    while(k > 0) {
      if(k & 1) B *= *this;
      *this *= *this;
      k >>= 1LL;
    }
    A.swap(B.A);
    return (*this);
  }

  Matrix operator+(const Matrix &B) const {
    return (Matrix(*this) += B);
  }

  Matrix operator-(const Matrix &B) const {
    return (Matrix(*this) -= B);
  }

  Matrix operator*(const Matrix &B) const {
    return (Matrix(*this) *= B);
  }

  Matrix operator^(const long long k) const {
    return (Matrix(*this) ^= k);
  }

  friend ostream &operator<<(ostream &os, Matrix &p) {
    size_t n = p.height(), m = p.width();
    for(int i = 0; i < n; i++) {
      os << "[";
      for(int j = 0; j < m; j++) {
        os << p[i][j] << (j + 1 == m ? "]\n" : ",");
      }
    }
    return (os);
  }


  T determinant() {
    Matrix B(*this);
    assert(width() == height());
    T ret = 1;
    for(int i = 0; i < width(); i++) {
      int idx = -1;
      for(int j = i; j < width(); j++) {
        if(B[j][i] != 0) idx = j;
      }
      if(idx == -1) return (0);
      if(i != idx) {
        ret *= -1;
        swap(B[i], B[idx]);
      }
      ret *= B[i][i];
      T vv = B[i][i];
      for(int j = 0; j < width(); j++) {
        B[i][j] /= vv;
      }
      for(int j = i + 1; j < width(); j++) {
        T a = B[j][i];
        for(int k = 0; k < width(); k++) {
          B[j][k] -= B[i][k] * a;
        }
      }
    }
    return (ret);
  }
};

template< int mod >
struct ModInt {
  int x;

  ModInt() : x(0) {}

  ModInt(int64_t y) : x(y >= 0 ? y % mod : (mod - (-y) % mod) % mod) {}

  ModInt &operator+=(const ModInt &p) {
    if((x += p.x) >= mod) x -= mod;
    return *this;
  }

  ModInt &operator-=(const ModInt &p) {
    if((x += mod - p.x) >= mod) x -= mod;
    return *this;
  }

  ModInt &operator*=(const ModInt &p) {
    x = (int) (1LL * x * p.x % mod);
    return *this;
  }

  ModInt &operator/=(const ModInt &p) {
    *this *= p.inverse();
    return *this;
  }

  ModInt operator-() const { return ModInt(-x); }

  ModInt operator+(const ModInt &p) const { return ModInt(*this) += p; }

  ModInt operator-(const ModInt &p) const { return ModInt(*this) -= p; }

  ModInt operator*(const ModInt &p) const { return ModInt(*this) *= p; }

  ModInt operator/(const ModInt &p) const { return ModInt(*this) /= p; }

  bool operator==(const ModInt &p) const { return x == p.x; }

  bool operator!=(const ModInt &p) const { return x != p.x; }

  ModInt inverse() const {
    int a = x, b = mod, u = 1, v = 0, t;
    while(b > 0) {
      t = a / b;
      swap(a -= t * b, b);
      swap(u -= t * v, v);
    }
    return ModInt(u);
  }

  ModInt pow(int64_t n) const {
    ModInt ret(1), mul(x);
    while(n > 0) {
      if(n & 1) ret *= mul;
      mul *= mul;
      n >>= 1;
    }
    return ret;
  }

  friend ostream &operator<<(ostream &os, const ModInt &p) {
    return os << p.x;
  }

  friend istream &operator>>(istream &is, ModInt &a) {
    int64_t t;
    is >> t;
    a = ModInt< mod >(t);
    return (is);
  }

  static int get_mod() { return mod; }
};

using modint = ModInt< mod >;

template< typename T >
struct FormalPowerSeries : vector< T > {
  using vector< T >::vector;
  using P = FormalPowerSeries;

  using MULT = function< P(P, P) >;
  using FFT = function< void(P &) >;

  static MULT &get_mult() {
    static MULT mult = nullptr;
    return mult;
  }

  static void set_mult(MULT f) { get_mult() = f; }

  static FFT &get_fft() {
    static FFT fft = nullptr;
    return fft;
  }

  static FFT &get_ifft() {
    static FFT ifft = nullptr;
    return ifft;
  }

  static void set_fft(FFT f, FFT g) {
    get_fft() = f;
    get_ifft() = g;
  }

  void shrink() {
    while(this->size() && this->back() == T(0)) this->pop_back();
  }

  P operator+(const P &r) const { return P(*this) += r; }

  P operator+(const T &v) const { return P(*this) += v; }

  P operator-(const P &r) const { return P(*this) -= r; }

  P operator-(const T &v) const { return P(*this) -= v; }

  P operator*(const P &r) const { return P(*this) *= r; }

  P operator*(const T &v) const { return P(*this) *= v; }

  P operator/(const P &r) const { return P(*this) /= r; }

  P operator%(const P &r) const { return P(*this) %= r; }

  P &operator+=(const P &r) {
    if(r.size() > this->size()) this->resize(r.size());
    for(int i = 0; i < r.size(); i++) (*this)[i] += r[i];
    return *this;
  }

  P &operator+=(const T &r) {
    if(this->empty()) this->resize(1);
    (*this)[0] += r;
    return *this;
  }

  P &operator-=(const P &r) {
    if(r.size() > this->size()) this->resize(r.size());
    for(int i = 0; i < r.size(); i++) (*this)[i] -= r[i];
    shrink();
    return *this;
  }

  P &operator-=(const T &r) {
    if(this->empty()) this->resize(1);
    (*this)[0] -= r;
    shrink();
    return *this;
  }

  P &operator*=(const T &v) {
    const int n = (int) this->size();
    for(int k = 0; k < n; k++) (*this)[k] *= v;
    return *this;
  }

  P &operator*=(const P &r) {
    if(this->empty() || r.empty()) {
      this->clear();
      return *this;
    }
    assert(get_mult() != nullptr);
    return *this = get_mult()(*this, r);
  }

  P &operator%=(const P &r) { return *this -= *this / r * r; }

  P operator-() const {
    P ret(this->size());
    for(int i = 0; i < this->size(); i++) ret[i] = -(*this)[i];
    return ret;
  }

  P &operator/=(const P &r) {
    if(this->size() < r.size()) {
      this->clear();
      return *this;
    }
    int n = this->size() - r.size() + 1;
    return *this = (rev().pre(n) * r.rev().inv(n)).pre(n).rev(n);
  }

  P pre(int sz) const { return P(begin(*this), begin(*this) + min((int) this->size(), sz)); }

  P operator>>(int sz) const {
    if(this->size() <= sz) return {};
    P ret(*this);
    ret.erase(ret.begin(), ret.begin() + sz);
    return ret;
  }

  P operator<<(int sz) const {
    P ret(*this);
    ret.insert(ret.begin(), sz, T(0));
    return ret;
  }

  P rev(int deg = -1) const {
    P ret(*this);
    if(deg != -1) ret.resize(deg, T(0));
    reverse(begin(ret), end(ret));
    return ret;
  }

  P diff() const {
    const int n = (int) this->size();
    P ret(max(0, n - 1));
    for(int i = 1; i < n; i++) ret[i - 1] = (*this)[i] * T(i);
    return ret;
  }

  P integral() const {
    const int n = (int) this->size();
    P ret(n + 1);
    ret[0] = T(0);
    for(int i = 0; i < n; i++) ret[i + 1] = (*this)[i] / T(i + 1);
    return ret;
  }

  // F(0) must not be 0
  P inv(int deg = -1) const {
    assert(((*this)[0]) != T(0));
    const int n = (int) this->size();
    if(deg == -1) deg = n;
    if(get_fft() != nullptr) {
      P ret(*this);
      ret.resize(deg, T(0));
      return ret.inv_fast();
    }
    P ret({T(1) / (*this)[0]});
    for(int i = 1; i < deg; i <<= 1) {
      ret = (ret + ret - ret * ret * pre(i << 1)).pre(i << 1);
    }
    return ret.pre(deg);
  }

  // F(0) must be 1
  P log(int deg = -1) const {
    assert((*this)[0] == 1);
    const int n = (int) this->size();
    if(deg == -1) deg = n;
    return (this->diff() * this->inv(deg)).pre(deg - 1).integral();
  }

  P sqrt(int deg = -1) const {
    const int n = (int) this->size();
    if(deg == -1) deg = n;
    if((*this)[0] == T(0)) {
      for(int i = 1; i < n; i++) {
        if((*this)[i] != T(0)) {
          if(i & 1) return {};
          if(deg - i / 2 <= 0) break;
          auto ret = (*this >> i).sqrt(deg - i / 2) << (i / 2);
          if(ret.size() < deg) ret.resize(deg, T(0));
          return ret;
        }
      }
      return P(deg, 0);
    }

    P ret({T(1)});
    T inv2 = T(1) / T(2);
    for(int i = 1; i < deg; i <<= 1) {
      ret = (ret + pre(i << 1) * ret.inv(i << 1)) * inv2;
    }
    return ret.pre(deg);
  }

  // F(0) must be 0
  P exp(int deg = -1) const {
    assert((*this)[0] == T(0));
    const int n = (int) this->size();
    if(deg == -1) deg = n;
    if(get_fft() != nullptr) {
      P ret(*this);
      ret.resize(deg, T(0));
      return ret.exp_rec();
    }
    P ret({T(1)});
    for(int i = 1; i < deg; i <<= 1) {
      ret = (ret * (pre(i << 1) + T(1) - ret.log(i << 1))).pre(i << 1);
    }
    return ret.pre(deg);
  }


  P online_convolution_exp(const P &conv_coeff) const {
    const int n = (int) conv_coeff.size();
    assert((n & (n - 1)) == 0);
    vector< P > conv_ntt_coeff;
    for(int i = n; i >= 1; i >>= 1) {
      P g(conv_coeff.pre(i));
      get_fft()(g);
      conv_ntt_coeff.emplace_back(g);
    }
    P conv_arg(n), conv_ret(n);
    auto rec = [&](auto rec, int l, int r, int d) -> void {
      if(r - l <= 16) {
        for(int i = l; i < r; i++) {
          T sum = 0;
          for(int j = l; j < i; j++) sum += conv_arg[j] * conv_coeff[i - j];
          conv_ret[i] += sum;
          conv_arg[i] = i == 0 ? T(1) : conv_ret[i] / i;
        }
      } else {
        int m = (l + r) / 2;
        rec(rec, l, m, d + 1);
        P pre(r - l);
        for(int i = 0; i < m - l; i++) pre[i] = conv_arg[l + i];
        get_fft()(pre);
        for(int i = 0; i < r - l; i++) pre[i] *= conv_ntt_coeff[d][i];
        get_ifft()(pre);
        for(int i = 0; i < r - m; i++) conv_ret[m + i] += pre[m + i - l];
        rec(rec, m, r, d + 1);
      }
    };
    rec(rec, 0, n, 0);
    return conv_arg;
  }

  P exp_rec() const {
    assert((*this)[0] == T(0));
    const int n = (int) this->size();
    int m = 1;
    while(m < n) m *= 2;
    P conv_coeff(m);
    for(int i = 1; i < n; i++) conv_coeff[i] = (*this)[i] * i;
    return online_convolution_exp(conv_coeff).pre(n);
  }


  P inv_fast() const {
    assert(((*this)[0]) != T(0));

    const int n = (int) this->size();
    P res{T(1) / (*this)[0]};

    for(int d = 1; d < n; d <<= 1) {
      P f(2 * d), g(2 * d);
      for(int j = 0; j < min(n, 2 * d); j++) f[j] = (*this)[j];
      for(int j = 0; j < d; j++) g[j] = res[j];
      get_fft()(f);
      get_fft()(g);
      for(int j = 0; j < 2 * d; j++) f[j] *= g[j];
      get_ifft()(f);
      for(int j = 0; j < d; j++) {
        f[j] = 0;
        f[j + d] = -f[j + d];
      }
      get_fft()(f);
      for(int j = 0; j < 2 * d; j++) f[j] *= g[j];
      get_ifft()(f);
      for(int j = 0; j < d; j++) f[j] = res[j];
      res = f;
    }
    return res.pre(n);
  }

  P pow(int64_t k, int deg = -1) const {
    const int n = (int) this->size();
    if(deg == -1) deg = n;
    for(int i = 0; i < n; i++) {
      if((*this)[i] != T(0)) {
        T rev = T(1) / (*this)[i];
        P ret = (((*this * rev) >> i).log() * k).exp() * ((*this)[i].pow(k));
        if(i * k > deg) return P(deg, T(0));
        ret = (ret << (i * k)).pre(deg);
        if(ret.size() < deg) ret.resize(deg, T(0));
        return ret;
      }
    }
    return *this;
  }

  T eval(T x) const {
    T r = 0, w = 1;
    for(auto &v : *this) {
      r += w * v;
      w *= x;
    }
    return r;
  }

  P pow_mod(int64_t n, P mod) {
    P x(*this), ret{1};
    while(n > 0) {
      if(n & 1) (ret *= x) %= mod;
      (x *= x) %= mod;
      n >>= 1;
    }
    return ret;
  }
};

template< typename Mint >
struct NumberTheoreticTransformFriendlyModInt {

  vector< int > rev;
  vector< Mint > rts;
  int base, max_base;
  Mint root;

  NumberTheoreticTransformFriendlyModInt() : base(1), rev{0, 1}, rts{0, 1} {
    const int mod = Mint::get_mod();
    assert(mod >= 3 && mod % 2 == 1);
    auto tmp = mod - 1;
    max_base = 0;
    while(tmp % 2 == 0) tmp >>= 1, max_base++;
    root = 2;
    while(root.pow((mod - 1) >> 1) == 1) root += 1;
    assert(root.pow(mod - 1) == 1);
    root = root.pow((mod - 1) >> max_base);
  }

  void ensure_base(int nbase) {
    if(nbase <= base) return;
    rev.resize(1 << nbase);
    rts.resize(1 << nbase);
    for(int i = 0; i < (1 << nbase); i++) {
      rev[i] = (rev[i >> 1] >> 1) + ((i & 1) << (nbase - 1));
    }
    assert(nbase <= max_base);
    while(base < nbase) {
      Mint z = root.pow(1 << (max_base - 1 - base));
      for(int i = 1 << (base - 1); i < (1 << base); i++) {
        rts[i << 1] = rts[i];
        rts[(i << 1) + 1] = rts[i] * z;
      }
      ++base;
    }
  }


  void ntt(vector< Mint > &a) {
    const int n = (int) a.size();
    assert((n & (n - 1)) == 0);
    int zeros = __builtin_ctz(n);
    ensure_base(zeros);
    int shift = base - zeros;
    for(int i = 0; i < n; i++) {
      if(i < (rev[i] >> shift)) {
        swap(a[i], a[rev[i] >> shift]);
      }
    }
    for(int k = 1; k < n; k <<= 1) {
      for(int i = 0; i < n; i += 2 * k) {
        for(int j = 0; j < k; j++) {
          Mint z = a[i + j + k] * rts[j + k];
          a[i + j + k] = a[i + j] - z;
          a[i + j] = a[i + j] + z;
        }
      }
    }
  }


  void intt(vector< Mint > &a) {
    const int n = (int) a.size();
    ntt(a);
    reverse(a.begin() + 1, a.end());
    Mint inv_sz = Mint(1) / n;
    for(int i = 0; i < n; i++) a[i] *= inv_sz;
  }

  vector< Mint > multiply(vector< Mint > a, vector< Mint > b) {
    int need = a.size() + b.size() - 1;
    int nbase = 1;
    while((1 << nbase) < need) nbase++;
    ensure_base(nbase);
    int sz = 1 << nbase;
    a.resize(sz, 0);
    b.resize(sz, 0);
    ntt(a);
    ntt(b);
    Mint inv_sz = Mint(1) / sz;
    for(int i = 0; i < sz; i++) {
      a[i] *= b[i] * inv_sz;
    }
    reverse(a.begin() + 1, a.end());
    ntt(a);
    a.resize(need);
    return a;
  }
};


template< typename T >
FormalPowerSeries< T > berlekamp_massey(const FormalPowerSeries< T > &s) {
  const int N = (int) s.size();
  FormalPowerSeries< T > B(N), C(N);
  B[0] = C[0] = T(1);
  int L = 0, m = 1;
  T b{1};
  for(int n = 0; n < N; n++) {
    T d = s[n];
    for(int i = 1; i <= L; i++) d += C[i] * s[n - i];
    T inv_b = d / b;
    if(d == 0) {
      m++;
    } else if(2 * L <= n) {
      auto temp = C;
      for(int i = 0; i + m < N; i++) C[i + m] -= B[i] * inv_b;
      L = n + 1 - L;
      B = temp;
      b = d;
      m = 1;
    } else {
      for(int i = 0; i + m < N; i++) C[i + m] -= B[i] * inv_b;
      m++;
    }
  }
  C.resize(L + 1);
  reverse(C.begin(), C.end());
  assert(L < N - 1);
  return C;
}

template< typename T >
FormalPowerSeries< T > random_poly(int n) {
  mt19937 mt(1333333);
  FormalPowerSeries< T > res(n);
  uniform_int_distribution< int > rand(0, T::get_mod() - 1);
  for(int i = 0; i < n; i++) res[i] = rand(mt);
  return res;
}

template< typename T, typename F >
FormalPowerSeries< T > minimum_poly(int N, const F &nex) {
  auto dp = random_poly< T >(N);
  auto u = random_poly< T >(N);
  FormalPowerSeries< T > f(2 * N + 5);
  for(int i = 0; i < 2 * N + 5; i++) {
    for(int j = 0; j < N; j++) f[i] += u[j] * dp[j];
    dp = nex(dp);
  }
  return berlekamp_massey< T >(f);
}


template< typename T, typename F >
FormalPowerSeries< T > mat_mul(int N, int64_t Q, FormalPowerSeries< modint > dp, const F &nex) {
  auto A = FormalPowerSeries< T >({0, 1}).pow_mod(Q, minimum_poly< T >(N, nex));
  FormalPowerSeries< T > res(N);
  for(int i = 0; i < A.size(); i++) {
    res += dp * A[i];
    dp = nex(dp);
  }
  return res;
}

int main() {
  using FPS = FormalPowerSeries< modint >;
  NumberTheoreticTransformFriendlyModInt< modint > fft;
  FPS::set_fft([&](FPS &a) { fft.ntt(a); }, [&](FPS &a) { fft.intt(a); });
  FPS::set_mult([&](const FPS &a, const FPS &b) {
    auto c = fft.multiply(a, b);
    return FPS(begin(c), end(c));
  });

  int N, K;
  cin >> N >> K;
  Matrix< modint > uku(K * K * K);
  auto to_idx = [&](int x, int y, int z) {
    return x * K * K + y * K + z;
  };
  for(int i = 0; i < K; i++) { // ほ
    for(int j = 0; j < K; j++) { // ほむ
      for(int k = 0; k < K; k++) { // ほむら

        // ほ
        uku[to_idx(i, j, k)][to_idx((i + 1) % K, j, k)] += 1;

        // む
        uku[to_idx(i, j, k)][to_idx(i, (i + j) % K, k)] += 1;

        // ら
        uku[to_idx(i, j, k)][to_idx(i, j, (j + k) % K)] += 1;
      }
    }
  }
  vector< pair< int, int > > ord;
  for(int i = 0; i < K * K * K; i++) {
    for(int j = 0; j < K * K * K; j++) {
      if(uku[i][j].x != 0) {
        ord.emplace_back(i, j);
      }
    }
  }

  FPS dp(K * K * K);
  dp[0] = 1;
  auto nex = [&](const FPS &a) {
    FPS b(a.size());
    for(auto &p : ord) b[p.second] += a[p.first] * uku[p.first][p.second];
    return b;
  };
  auto mat = mat_mul< modint >(K * K * K, N, dp, nex);
  modint ret = 0;
  for(int i = 0; i < K; i++) {
    for(int j = 0; j < K; j++) {
      ret += mat[to_idx(i, j, 0)];
    }
  }
  cout << ret << endl;
}