結果
問題 | No.2074 Product is Square ? |
ユーザー | YFuka86 |
提出日時 | 2022-09-16 22:32:41 |
言語 | C++17 (gcc 12.3.0 + boost 1.83.0) |
結果 |
AC
|
実行時間 | 166 ms / 2,000 ms |
コード長 | 21,897 bytes |
コンパイル時間 | 3,881 ms |
コンパイル使用メモリ | 258,384 KB |
実行使用メモリ | 6,944 KB |
最終ジャッジ日時 | 2024-06-01 13:36:11 |
合計ジャッジ時間 | 6,609 ms |
ジャッジサーバーID (参考情報) |
judge1 / judge4 |
(要ログイン)
テストケース
テストケース表示入力 | 結果 | 実行時間 実行使用メモリ |
---|---|---|
testcase_00 | AC | 2 ms
6,816 KB |
testcase_01 | AC | 38 ms
6,816 KB |
testcase_02 | AC | 39 ms
6,940 KB |
testcase_03 | AC | 40 ms
6,944 KB |
testcase_04 | AC | 39 ms
6,940 KB |
testcase_05 | AC | 41 ms
6,940 KB |
testcase_06 | AC | 41 ms
6,940 KB |
testcase_07 | AC | 40 ms
6,940 KB |
testcase_08 | AC | 41 ms
6,944 KB |
testcase_09 | AC | 39 ms
6,944 KB |
testcase_10 | AC | 40 ms
6,940 KB |
testcase_11 | AC | 16 ms
6,940 KB |
testcase_12 | AC | 42 ms
6,944 KB |
testcase_13 | AC | 162 ms
6,940 KB |
testcase_14 | AC | 27 ms
6,944 KB |
testcase_15 | AC | 15 ms
6,944 KB |
testcase_16 | AC | 44 ms
6,944 KB |
testcase_17 | AC | 166 ms
6,944 KB |
testcase_18 | AC | 29 ms
6,944 KB |
testcase_19 | AC | 14 ms
6,940 KB |
testcase_20 | AC | 43 ms
6,944 KB |
testcase_21 | AC | 159 ms
6,944 KB |
testcase_22 | AC | 27 ms
6,944 KB |
testcase_23 | AC | 15 ms
6,944 KB |
testcase_24 | AC | 43 ms
6,944 KB |
testcase_25 | AC | 160 ms
6,940 KB |
testcase_26 | AC | 29 ms
6,940 KB |
testcase_27 | AC | 15 ms
6,940 KB |
testcase_28 | AC | 47 ms
6,940 KB |
testcase_29 | AC | 163 ms
6,940 KB |
testcase_30 | AC | 27 ms
6,944 KB |
testcase_31 | AC | 5 ms
6,940 KB |
testcase_32 | AC | 2 ms
6,940 KB |
testcase_33 | AC | 3 ms
6,940 KB |
ソースコード
#pragma GCC optimize("Ofast") #include <bits/stdc++.h> using namespace std; using ll = long long; using ull = unsigned long long; using ld = long double; using i128 = __int128_t; using P = pair<int, int>; using LP = pair<ll, ll>; using LT = tuple<ll, ll, ll>; using LT4 = tuple<ll, ll, ll, ll>; typedef vector<int> vi; typedef vector<vi> vvi; typedef vector<ll> vl; typedef vector<vl> vvl; typedef vector<vvl> v3l; typedef vector<v3l> v4l; typedef vector<v4l> v5l; typedef vector<LP> vlp; typedef vector<vlp> vvlp; typedef vector<LT> vlt; typedef vector<vlt> vvlt; typedef vector<LT4> vlt4; typedef vector<string> vs; typedef vector<vs> vvs; typedef vector<ld> vd; typedef vector<vd> vvd; typedef vector<bool> vb; typedef vector<vb> vvb; template<typename T> using pq = priority_queue<T>; template<typename T> using mpq = priority_queue<T, vector<T>, greater<T>>; template<typename T> class infinity{ public: static constexpr T MAX=numeric_limits<T>::max(); static constexpr T MIN=numeric_limits<T>::min(); static constexpr T val=numeric_limits<T>::max()/2-1e6; static constexpr T mval=numeric_limits<T>::min()/2+1e6; }; const int INF = infinity<int>::val; const ll LINF = infinity<ll>::val; const ld DINF = infinity<ld>::val; #define _overload5(a, b, c, d, e, name, ...) name #define _overload4(a, b, c, d, name, ...) name #define _overload3(a, b, c, name, ...) name #define _rep0(n) for(ll i = 0; (ll)(i) < n; ++i) #define _rep1(i, n) for(ll i = 0; i < (ll)(n); ++i) #define _rep2(i, a, b) for(ll i = (ll)(a); i < (ll)(b); ++i) #define _rep3(i, a, b, c) for(ll i = (ll)(a); i < (ll)(b); i += (ll)(c)) #define rep(...) _overload4(__VA_ARGS__, _rep3, _rep2, _rep1, _rep0)(__VA_ARGS__) #define _per0(n) for(ll i = 0; i < (ll)(n); ++i) #define _per1(i, n) for(ll i = (ll)(n)-1; i >= 0; --i) #define _per2(i, a, b) for(ll i = (ll)(a)-1; i >= (ll)(b); --i) #define _per3(i, a, b, c) for(ll i = (ll)(a)-1; i >= (ll)(b); i -= (ll)(c)) #define rep_r(...) _overload4(__VA_ARGS__, _per3, _per2, _per1, _per0)(__VA_ARGS__) #define _fore0(a) rep(a.size()) #define _fore1(i, a) for(auto &&i : a) #define _fore2(a, b, v) for(auto &&[a, b] : v) #define _fore3(a, b, c, v) for(auto &&[a, b, c] : v) #define _fore4(a, b, c, d, v) for(auto &&[a, b, c, d] : v) #define fore(...) _overload5(__VA_ARGS__, _fore4, _fore3, _fore2, _fore1, _fore0)(__VA_ARGS__) #define all(v) (v).begin(),(v).end() #define rall(v) (v).rbegin(),(v).rend() #define RNG(v, l, r) (v).begin() + l, (v).begin() + r #define pb push_back #define eb emplace_back #define mp make_pair #define fi first #define se second #define vv(type, name, h, ...) vector<vector<type>> name(h, vector<type>(__VA_ARGS__)) #define v3(type, name, h, w, ...) vector<vector<vector<type>>> name(h, vector<vector<type>>(w, vector<type>(__VA_ARGS__))) #define v4(type, name, a, b, c, ...) vector<vector<vector<vector<type>>>> name(a, vector<vector<vector<type>>>(b, vector<vector<type>>(c, vector<type>(__VA_ARGS__)))) struct RandGen { using ud = uniform_int_distribution<ll>; mt19937 mt; RandGen() : mt(chrono::steady_clock::now().time_since_epoch().count()) {} ll l(ll a, ll b) { ud d(a, b - 1); return d(mt); } LP lp(ll a, ll b, bool rng = true) { ll x = l(a, b - 1), y = l(rng ? x + 1 : a, b - 1); return {x, y}; } vl vecl(ll n, ll a, ll b) { ud d(a, b - 1); vl ret(n); rep(i, n) ret[i] = d(mt); return ret; } vl vecperm(ll n, ll from = 0) { vl perm(n); iota(all(perm), from); shuffle(perm); return perm; } string str(ll n, string op) { vl fig = vecl(n, 0, op.size()); string s; rep(i, n) s.pb(op[fig[i]]); return s; } string straz(ll n, ll a = 0, ll z = 26) { vl az = vecl(n, a, z); string s; rep(i, n) s.pb('a' + az[i]); return s; } string strnum(ll n, ll zero = 0, ll ten = 10) { vl zt = vecl(n, zero, ten); string s; rep(i, n) s.pb('0' + zt[i]); return s; } template<typename T> void shuffle(vector<T> &a) { std::shuffle(all(a), mt); } }; // デバッグ系 #define dout cout template<typename T, typename=void> struct is_specialize:false_type{}; template<typename T> struct is_specialize<T, typename conditional<false,typename T::iterator, void>::type>:true_type{}; template<typename T> struct is_specialize<T, typename conditional<false,decltype(T::first), void>::type>:true_type{}; template<typename T> struct is_specialize<T, enable_if_t<is_integral<T>::value, void>>:true_type{}; void dump(const char &t) { dout<<t; } void dump(const string &t) { dout<<t; } void dump(const bool &t) { dout<<(t ? "true" : "false"); } template<typename T, enable_if_t<!is_specialize<T>::value, nullptr_t> =nullptr> void dump(const T&t) { dout << t; } template<typename T> void dump(const T&t, enable_if_t<is_integral<T>::value>* =nullptr) { string tmp;if(t==infinity<T>::val||t==infinity<T>::MAX)tmp="inf";if(is_signed<T>::value&&(t==infinity<T>::mval||t==infinity<T>::MIN))tmp="-inf";if(tmp.empty())tmp=to_string(t);dout<<tmp; } template<typename T, typename U, typename V> void dump(const tuple<T, U, V>&t) { dout<<"("; dump(get<0>(t)); dout<<" "; dump(get<1>(t)); dout << " "; dump(get<2>(t)); dout << ")"; } template<typename T,typename U> void dump(const pair<T,U>&); template<typename T> void dump(const T&t, enable_if_t<!is_void<typename T::iterator>::value>* =nullptr) { dout << "{ "; for(auto it=t.begin();it!=t.end();){ dump(*it); dout << (++it==t.end() ? "" : " "); } dout<<" }"; } template<typename T,typename U> void dump(const pair<T,U>&t) { dout<<"("; dump(t.first); dout<<" "; dump(t.second); dout << ")"; } void trace() { dout << "\n"; } template<typename Head, typename... Tail> void trace(Head&&head, Tail&&... tail) { dump(head); if(sizeof...(tail)) dout<<", "; trace(forward<Tail>(tail)...); } #ifdef ONLINE_JUDGE #define debug(...) (void(0)) #else #define debug(...) do {dout<<#__VA_ARGS__<<" = ";trace(__VA_ARGS__); } while(0) #endif // 入出力系 #define LL(...) ll __VA_ARGS__; IN(__VA_ARGS__) #define STR(...) string __VA_ARGS__; IN(__VA_ARGS__) #define LD(...) ld __VA_ARGS__; IN(__VA_ARGS__) #define _vl(name, size) vl name(size); IN(name) #define _vl2(name, size, off) vl name(size); IN(name); rep(i, size) name[i]-=off #define VL(...) _overload3(__VA_ARGS__, _vl2, _vl)(__VA_ARGS__) #define VEC(type, name, size) vector<type> name(size); IN(name) #define VEC2(type, name1, name2, size) vector<type> name1(size), name2(size); for(int i = 0; i < size; i++) IN(name1[i], name2[i]) #define VEC3(type, name1, name2, name3, size) vector<type> name1(size), name2(size), name3(size); for(int i = 0; i < size; i++) IN(name1[i], name2[i], name3[i]) #define VEC4(type, name1, name2, name3, name4, size) vector<type> name1(size), name2(size), name3(size), name4(size); for(int i = 0; i < size; i++) IN(name1[i], name2[i], name3[i], name4[i]); #define VV(type, name, h, w) vector<vector<type>> name(h, vector<type>(w)); IN(name) void scan(int &a) { cin >> a; } void scan(long long &a) { cin >> a; } void scan(char &a) { cin >> a; } void scan(double &a) { cin >> a; } void scan(string &a) { cin >> a; } template <class T, class S> void scan(pair<T, S> &p) { scan(p.first), scan(p.second); } template <class T> void scan(vector<T> &a) { for(auto &i : a) scan(i); } template <class T> void scan(T &a) { cin >> a; } void IN() {} template <class Head, class... Tail> void IN(Head &head, Tail &...tail) { scan(head); IN(tail...); } #define OUTRET(...) { { OUT(__VA_ARGS__); return; } } template <class T, class S> ostream &operator<<(ostream &os, const pair<T, S> &p) { return os << p.first << " " << p.second; } void OUT() { cout << '\n'; } template <typename Head, typename... Tail> void OUT(const Head &head, const Tail &...tail) { cout << head; if(sizeof...(tail)) cout << ' '; OUT(tail...); } template<typename T> void OUTARRAY(vector<T>& v, int offset = 0, string sep = " ") { rep(i, v.size()) { if (i > 0) cout << sep; if (offset) cout << v[i] + offset; else cout << v[i]; } cout << "\n"; } template<typename T> void OUTMAT(vector<vector<T>>& v, int offset = 0) { rep(i, v.size()) { coutarray(v[i], offset); } } template<typename T> void OUTBIN(T &a, int d) { for (int i = d - 1; i >= 0; i--) cout << ((a >> i) & (T)1); cout << "\n"; } template<typename Q, typename A> void IQUERY(initializer_list<Q> q, A &a, string str = "? ") { cout << str; vector<Q> v(q); coutarray(v); cout.flush(); cin >> a; } // template<typename Q, typename A> void IQUERY(initializer_list<Q> q, A &a, string str = "? ") { vector<Q> query(q); RandGen rg; // a = query[0] ? A() : A(); // } template<typename A> void IANSWER(A a, string str = "! ") { cout << str << a << "\n"; cout.flush(); } // 数値系 int ceil_pow2(ll n) { int x = 0; while ((1ULL << x) < (ull)(n)) x++; return x; } int floor_pow2(ll n) { int x = 0; while ((1ULL << (x + 1)) <= (ull)(n)) x++; return x; } ll POW(__uint128_t x, int n) { assert(n >= 0); ll res = 1; for(; n; n >>= 1, x *= x) if(n & 1) res *= x; return res; } ll sqrt_ceil(ll x) { ll l = -1, r = x; while (r - l > 1) { ll m = (l + r) / 2; if (m * m >= x) r = m; else l = m; } return r; } vl divisor(ll n) { vl ret; for (ll i = 1; i * i <= n; i++) { if (n % i == 0) { ret.pb(i); if (i * i != n) ret.pb(n / i); } } sort(all(ret)); return ret; } template<typename T> vl digits(T n) { assert(n >= 0); vl ret; while(n > 0) { ret.pb(n % 10); n /= 10; } return ret; } template<typename T, typename S> T ceil(T x, S y) { assert(y); return (y < 0 ? ceil(-x, -y) : (x > 0 ? (x + y - 1) / y : x / y)); } template<typename T, typename S> T floor(T x, S y) { assert(y); return (y < 0 ? floor(-x, -y) : (x > 0 ? x / y : (x - y + 1) / y)); } template<typename T = ll> T sum_of(const vector<T> &v, int l = 0, int r = INF) { return accumulate(rng(v, l, min(r, (int)v.size())), T(0)); } ll max(int x, ll y) { return max((ll)x, y); } ll max(ll x, int y) { return max(x, (ll)y); } ll min(int x, ll y) { return min((ll)x, y); } ll min(ll x, int y) { return min(x, (ll)y); } ll mex(vl& v) { ll n = v.size(); vb S(n + 1); for (auto a: v) if (a <= n) S[a] = 1; ll ret = 0; while (S[ret]) ret++; return ret; } // 操作系 template<typename T> void uniq(vector<T>&a) { sort(all(a)); a.erase(unique(all(a)), a.end()); } template<typename T> void comp(vector<T>&a) { vector<T> b = a; uniq(b); rep(i, a.size()) a[i] = lower_bound(all(b), a[i]) - b.begin(); } template<class T, class U> bool chmin(T &a, const U &b) { if (b < a) { a = b; return 1;} return 0; } template<class T, class U> bool chmax(T &a, const U &b) { if (b > a) { a = b; return 1;} return 0; } template<class T> int lbs(vector<T> &a, const T &b) { return lower_bound(all(a), b) - a.begin(); }; template<class T> int ubs(vector<T> &a, const T &b) { return upper_bound(all(a), b) - a.begin(); }; ll binary_search(function<bool(ll)> check, ll ok, ll ng) { assert(check(ok)); while (abs(ok - ng) > 1) { auto x = (ng + ok) / 2; if (check(x)) ok = x; else ng = x; } return ok; } template<class T> vector<T> csum(vector<T> &a) { vl ret(a.size() + 1, 0); rep(i, a.size()) ret[i + 1] = ret[i] + a[i]; return ret; } template<class S> vector<pair<S, int>> RLE(const vector<S> &v) { vector<pair<S, int>> res; for(auto &e : v) if(res.empty() or res.back().first != e) res.emplace_back(e, 1); else res.back().second++; return res; } vector<pair<char, int>> RLE(const string &v) { vector<pair<char, int>> res; for(auto &e : v) if(res.empty() or res.back().first != e) res.emplace_back(e, 1); else res.back().second++; return res; } template <class T, class S> bool incl(const T &x, const S &l, const S &r) { return l <= x and x < r; } bool is_palindrome(string s) { rep(i, (s.size() + 1) / 2) if (s[i] != s[s.size() - 1 - i]) { return false; } return true; } const string drul = "DRUL"; vl dx = {1, 0, -1, 0}; vl dy = {0, 1, 0, -1}; ll solve(ll n, vl a) { ll ans = n - a[0]; return ans; } ll naive(ll n, vl a) { ll ans = n + a[0]; return ans; } void compare(bool check = true) { RandGen rg; ll c = 0, loop = 10; while (++c) { if (c % loop == 0) cout << "reached " << c / loop << "loop" << "\n", cout.flush(); ll n = 10; vl a = rg.vecl(n, 1, 1e2); auto so = solve(n, a); auto na = naive(n, a); if (!check || na != so) { cout << c << "times tried" << "\n"; debug(n, a); debug(so); debug(na); if (check || (!check && c > loop)) break; } } } // https://nyaannyaan.github.io/library/inner/inner_math.hpp namespace inner { using i32 = int32_t; using u32 = uint32_t; using i64 = int64_t; using u64 = uint64_t; template <typename T> T gcd(T a, T b) { while (b) swap(a %= b, b); return a; } template <typename T> T inv(T a, T p) { T b = p, x = 1, y = 0; while (a) { T q = b / a; swap(a, b %= a); swap(x, y -= q * x); } assert(b == 1); return y < 0 ? y + p : y; } template <typename T, typename U> T modpow(T a, U n, T p) { T ret = 1 % p; for (; n; n >>= 1, a = U(a) * a % p) if (n & 1) ret = U(ret) * a % p; return ret; } } // namespace inner // https://nyaannyaan.github.io/library/misc/rng.hpp namespace my_rand { // [0, 2^64 - 1) uint64_t rng() { static uint64_t x_ = uint64_t(chrono::duration_cast<chrono::nanoseconds>(chrono::high_resolution_clock::now().time_since_epoch()).count()) * 10150724397891781847ULL; x_ ^= x_ << 7; return x_ ^= x_ >> 9; } // [l, r) int64_t randint(int64_t l, int64_t r) { assert(l < r); return l + rng() % (r - l); } // choose n numbers from [l, r) without overlapping vector<int64_t> randset(int64_t l, int64_t r, int64_t n) { assert(l <= r && n <= r - l); unordered_set<int64_t> s; for (int64_t i = n; i; --i) { int64_t m = randint(l, r + 1 - i); if (s.find(m) != s.end()) m = r - i; s.insert(m); } vector<int64_t> ret; for (auto& x : s) ret.push_back(x); return ret; } // [0.0, 1.0) double rnd() { union raw_cast { double t; uint64_t u; }; constexpr uint64_t p = uint64_t(1023 - 64) << 52; return rng() * ((raw_cast*)(&p))->t; } template <typename T> void randshf(vector<T>& v) { int n = v.size(); for (int loop = 0; loop < 2; loop++) for (int i = 0; i < n; i++) swap(v[i], v[randint(0, n)]); } } // namespace my_rand using my_rand::randint; using my_rand::randset; using my_rand::randshf; using my_rand::rnd; using my_rand::rng; // https://nyaannyaan.github.io/library/modint/arbitrary-prime-modint.hpp struct ArbitraryLazyMontgomeryModInt { using mint = ArbitraryLazyMontgomeryModInt; using i32 = int32_t; using u32 = uint32_t; using u64 = uint64_t; static u32 mod; static u32 r; static u32 n2; static u32 get_r() { u32 ret = mod; for (i32 i = 0; i < 4; ++i) ret *= 2 - mod * ret; return ret; } static void set_mod(u32 m) { assert(m < (1 << 30)); assert((m & 1) == 1); mod = m; n2 = -u64(m) % m; r = get_r(); assert(r * mod == 1); } u32 a; ArbitraryLazyMontgomeryModInt() : a(0) {} ArbitraryLazyMontgomeryModInt(const int64_t &b): a(reduce(u64(b % mod + mod) * n2)){}; static u32 reduce(const u64 &b) { return (b + u64(u32(b) * u32(-r)) * mod) >> 32; } mint &operator+=(const mint &b) { if (i32(a += b.a - 2 * mod) < 0) a += 2 * mod; return *this; } mint &operator-=(const mint &b) { if (i32(a -= b.a) < 0) a += 2 * mod; return *this; } mint &operator*=(const mint &b) { a = reduce(u64(a) * b.a); return *this; } mint &operator/=(const mint &b) { *this *= b.inverse(); return *this; } mint operator+(const mint &b) const { return mint(*this) += b; } mint operator-(const mint &b) const { return mint(*this) -= b; } mint operator*(const mint &b) const { return mint(*this) *= b; } mint operator/(const mint &b) const { return mint(*this) /= b; } bool operator==(const mint &b) const { return (a >= mod ? a - mod : a) == (b.a >= mod ? b.a - mod : b.a); } bool operator!=(const mint &b) const { return (a >= mod ? a - mod : a) != (b.a >= mod ? b.a - mod : b.a); } mint operator-() const { return mint() - mint(*this); } mint pow(u64 n) const { mint ret(1), mul(*this); while (n > 0) { if (n & 1) ret *= mul; mul *= mul; n >>= 1; } return ret; } friend ostream &operator<<(ostream &os, const mint &b) { return os << b.get(); } friend istream &operator>>(istream &is, mint &b) { int64_t t; is >> t; b = ArbitraryLazyMontgomeryModInt(t); return (is); } mint inverse() const { return pow(mod - 2); } u32 get() const { u32 ret = reduce(a); return ret >= mod ? ret - mod : ret; } static u32 get_mod() { return mod; } }; typename ArbitraryLazyMontgomeryModInt::u32 ArbitraryLazyMontgomeryModInt::mod; typename ArbitraryLazyMontgomeryModInt::u32 ArbitraryLazyMontgomeryModInt::r; typename ArbitraryLazyMontgomeryModInt::u32 ArbitraryLazyMontgomeryModInt::n2; // https://nyaannyaan.github.io/library/modint/modint-montgomery64.hpp struct montgomery64 { using mint = montgomery64; using i64 = int64_t; using u64 = uint64_t; using u128 = __uint128_t; static u64 mod; static u64 r; static u64 n2; static u64 get_r() { u64 ret = mod; for (i64 i = 0; i < 5; ++i) ret *= 2 - mod * ret; return ret; } static void set_mod(u64 m) { assert(m < (1LL << 62)); assert((m & 1) == 1); mod = m; n2 = -u128(m) % m; r = get_r(); assert(r * mod == 1); } u64 a; montgomery64() : a(0) {} montgomery64(const int64_t &b) : a(reduce((u128(b) + mod) * n2)){}; static u64 reduce(const u128 &b) { return (b + u128(u64(b) * u64(-r)) * mod) >> 64; } mint &operator+=(const mint &b) { if (i64(a += b.a - 2 * mod) < 0) a += 2 * mod; return *this; } mint &operator-=(const mint &b) { if (i64(a -= b.a) < 0) a += 2 * mod; return *this; } mint &operator*=(const mint &b) { a = reduce(u128(a) * b.a); return *this; } mint &operator/=(const mint &b) { *this *= b.inverse(); return *this; } mint operator+(const mint &b) const { return mint(*this) += b; } mint operator-(const mint &b) const { return mint(*this) -= b; } mint operator*(const mint &b) const { return mint(*this) *= b; } mint operator/(const mint &b) const { return mint(*this) /= b; } bool operator==(const mint &b) const { return (a >= mod ? a - mod : a) == (b.a >= mod ? b.a - mod : b.a); } bool operator!=(const mint &b) const { return (a >= mod ? a - mod : a) != (b.a >= mod ? b.a - mod : b.a); } mint operator-() const { return mint() - mint(*this); } mint pow(u128 n) const { mint ret(1), mul(*this); while (n > 0) { if (n & 1) ret *= mul; mul *= mul; n >>= 1; } return ret; } friend ostream &operator<<(ostream &os, const mint &b) { return os << b.get(); } friend istream &operator>>(istream &is, mint &b) { int64_t t; is >> t; b = montgomery64(t); return (is); } mint inverse() const { return pow(mod - 2); } u64 get() const { u64 ret = reduce(a); return ret >= mod ? ret - mod : ret; } static u64 get_mod() { return mod; } }; typename montgomery64::u64 montgomery64::mod, montgomery64::r, montgomery64::n2; // https://nyaannyaan.github.io/library/prime/fast-factorize.hpp namespace fast_factorize { using u64 = uint64_t; template <typename mint> bool miller_rabin(u64 n, vector<u64> as) { if (mint::get_mod() != n) mint::set_mod(n); u64 d = n - 1; while (~d & 1) d >>= 1; mint e{1}, rev{int64_t(n - 1)}; for (u64 a : as) { if (n <= a) break; u64 t = d; mint y = mint(a).pow(t); while (t != n - 1 && y != e && y != rev) { y *= y; t *= 2; } if (y != rev && t % 2 == 0) return false; } return true; } bool is_prime(u64 n) { if (~n & 1) return n == 2; if (n <= 1) return false; if (n < (1LL << 30)) return miller_rabin<ArbitraryLazyMontgomeryModInt>(n, {2, 7, 61}); else return miller_rabin<montgomery64>( n, {2, 325, 9375, 28178, 450775, 9780504, 1795265022}); } template <typename mint, typename T> T pollard_rho(T n) { if (~n & 1) return 2; if (is_prime(n)) return n; if (mint::get_mod() != n) mint::set_mod(n); mint R, one = 1; auto f = [&](mint x) { return x * x + R; }; auto rnd_ = [&]() { return rng() % (n - 2) + 2; }; while (1) { mint x, y, ys, q = one; R = rnd_(), y = rnd_(); T g = 1; constexpr int m = 128; for (int r = 1; g == 1; r <<= 1) { x = y; for (int i = 0; i < r; ++i) y = f(y); for (int k = 0; g == 1 && k < r; k += m) { ys = y; for (int i = 0; i < m && i < r - k; ++i) q *= x - (y = f(y)); g = inner::gcd<T>(q.get(), n); } } if (g == n) do g = inner::gcd<T>((x - (ys = f(ys))).get(), n); while (g == 1); if (g != n) return g; } exit(1); } using i64 = long long; vector<i64> inner_factorize(u64 n) { if (n <= 1) return {}; u64 p; if (n <= (1LL << 30)) p = pollard_rho<ArbitraryLazyMontgomeryModInt, uint32_t>(n); else p = pollard_rho<montgomery64, uint64_t>(n); if (p == n) return {i64(p)}; auto l = inner_factorize(p); auto r = inner_factorize(n / p); copy(begin(r), end(r), back_inserter(l)); return l; } vector<i64> factorize(u64 n) { auto ret = inner_factorize(n); sort(begin(ret), end(ret)); return ret; } map<i64, i64> factor_count(u64 n) { map<i64, i64> mp; for (auto &x : factorize(n)) mp[x]++; return mp; } vector<i64> divisors(u64 n) { if (n == 0) return {}; vector<pair<i64, i64>> v; for (auto &p : factorize(n)) { if (v.empty() || v.back().first != p) { v.emplace_back(p, 1); } else { v.back().second++; } } vector<i64> ret; auto f = [&](auto rc, int i, i64 x) -> void { if (i == (int)v.size()) { ret.push_back(x); return; } for (int j = v[i].second;; --j) { rc(rc, i + 1, x); if (j == 0) break; x *= v[i].first; } }; f(f, 0, 1); sort(begin(ret), end(ret)); return ret; } } // namespace fast_factorize using fast_factorize::divisors; using fast_factorize::factor_count; using fast_factorize::factorize; using fast_factorize::is_prime; void solve() { LL(n); VL(a, n); map<ll, ll> cnt; rep(i, n) { fore(p, factorize(a[i])) { cnt[p]++; } } fore(_,c,cnt) { if (c & 1) OUTRET("No"); } OUT("Yes"); } signed main() { cin.tie(0)->sync_with_stdio(0); cout.tie(0); cout << fixed << setprecision(20); int t; cin >> t; while (t--) solve(); // while (t--) compare(); }