void DecideSearchingMethod()
{
if(linear())
	Binary_Search();
else
	Ternary_Search();
}

public boolean PathAvoidKOZ(Point begin, Point end)
    {
        for(int i=0;i<KOZ_MAX.size();i++)
        {
            if((begin.getX()>=KOZ_MIN.get(i).getX()&&begin.getX()<=KOZ_MAX.get(i).getX())||end.getX()<=KOZ_MAX.get(i).getX()&&end.getX()>=KOZ_MIN.get(i).getX())
                if((begin.getY()>=KOZ_MIN.get(i).getY()&&begin.getY()<=KOZ_MAX.get(i).getY())||end.getY()<=KOZ_MAX.get(i).getY()&&end.getY()>=KOZ_MIN.get(i).getY())
                    if((begin.getZ()>=KOZ_MIN.get(i).getZ()&&begin.getZ()<=KOZ_MAX.get(i).getZ())||end.getZ()<=KOZ_MAX.get(i).getZ()&&end.getZ()>=KOZ_MIN.get(i).getZ())
                        return true;
        }
        return false;
    }

#include<bits/stdc++.h>
using namespace std;
#define mx 200005

vector<int>g[mx];
int visited[mx];
stack<int>s;
void dfs(int u)
{
    visited[u]=1;
    for(int i=0; i<g[u].size(); i++)
    {
        int v=g[u][i];
        if(visited[v]==0) dfs(v);
    }
    s.push(u);
}
int main()
{
    int n;
    cin>>n; //number of vertex

    int m;
    cin>>m; //number of edges
    while(m--)
    {
        int u,v;
        cin>>u>>v;

        g[u].push_back(v); //making one directional graph

    }
    for(int i=1;i<=n;i++){
        if(visited[i]==0) dfs(i);
    }

    for(int i=0; i<n; i++)
    {    
        int x=s.top();
         x.pop();

        cout<<x<" ";

    }
    cout<<endl;

}

#include <stdio.h>
#include <string.h>

int main() {
    int t, v;
    char line[100], h[2];
    scanf("%d ", &t);
    while(t--) {
        int map[26][26] = {}, day[26] = {}, mt[26] = {};
        int i, j, done[26] = {};
        while(gets(line)) {
            if(line[0] == '\0')
                break;
            sscanf(line, "%s %d", h, &v);
            j = h[0]-'A', done[j] = 1;
            day[j] = v;
            for(i = strlen(line)-1; i >= 0; i--) {
                if(line[i] < 'A' || line[i] > 'Z')
                    break;
                map[j][mt[j]++] = line[i]-'A';
            }
        }
        int fin[26] = {}, used[26] = {};
        int max, ans = 0;
        while(1) {
            int flag = 0;
            for(i = 0; i < 26; i++) {
                if(used[i] == 0 && done[i] == 1) {
                    max = 0;
                    for(j = 0; j < mt[i]; j++) {
                        if(used[map[i][j]] == 0)
                            break;
                        if(fin[map[i][j]] > max)
                            max = fin[map[i][j]];
                    }
                    if(j == mt[i]) {
                        fin[i] = max+day[i];
                        if(fin[i] > ans)
                            ans = fin[i];
                        used[i] = 1;
                        flag = 1;
                    }
                }
            }
            if(flag == 0)
                break;
        }
        printf("%d\n", ans);
        if(t)
            puts("");
    }
    return 0;
}

# Basis
import matplotlib.pyplot as plt
import numpy as np
import qiskit as q
from math import pi
from qiskit import *

# Plot data
from qiskit.visualization import plot_histogram
from qiskit.tools.monitor import job_monitor

# Quantum Circuit
from qiskit import QuantumCircuit, ClassicalRegister, QuantumRegister, transpile
from qiskit.tools.visualization import circuit_drawer
from qiskit.quantum_info import state_fidelity

# Simulator
from qiskit import BasicAer

# Tell matplotlib that we are in an Ipython notebook
%matplotlib inline

# This creates a quantum circuit object
circ = q.QuantumCircuit(1,1)

# This prints out a diagram of the circuit, the "mpl" makes it look pretty!
display(circ.draw(output="mpl"))

# Linear operation to qubit 0
circ.h(0)
circ.z(0)
circ.h(0)
display(circ.draw(output="mpl"))

# Multi-qubit
circ = q.QuantumCircuit(15,1)
circ.x(6)
display(circ.draw(output="mpl"))

# rz gate
circ.rz(m.pi,0)
display(circ.draw(output="mpl"))

# Call Aer for simulation
backend=q.Aer.get_backend('statevector_simulator')
job = q.execute(circ, backend)
result = job.result()
counts = result.get_counts(circ)
graph = q.visualization.plot_histogram(counts)
display(graph)

# Bell state: Φ+
phi_plus = q.QuantumCircuit(2)
phi_plus.h(0)
phi_plus.cx(0,1)
display(phi_plus.draw(output="mpl")) 

# Bell state: Φ- (Use Built-in Function)
circ_phiminus = QuantumCircuit(2,2)
circ_phiminus.h(0)
circ_phiminus.draw('mpl')

# Build Quantum Circuit
circ_phiminus = QuantumCircuit(2,2)
circ_phiminus.h(0)
circ_phiminus.draw('mpl')
circ_phiminus.z(0) #Applying a Z gate to qubit 0
circ_phiminus.cx(0,1) #adding CNOT gate with qubit 0 as control and qubit 1 as target

circ_phiminus.measure([0,1],[0,1]) # Measure output --> Classical bit
circ_phiminus.draw('mpl')

backend = Aer.get_backend('qasm_simulator')  # specifying that we will use qasm simulator
job = execute(circ_phiminus, backend, shots = 1024)  # shots=1024 specifies that the circuit will be run 1024 times
result = job.result()
# getting the counts, i.e., the fraction of times the circuit gave all the possible results
counts = result.get_counts(circ_phiminus)  
graph = plot_histogram(counts)
display(graph)

from qiskit_ibm_runtime import QiskitRuntimeService

options = {
	'backend_name': 'ibmq_qasm_simulator'
}

runtime_inputs = {
	# A PyTorch-compatible loss function. Can be one of the official PyTorch loss 
	# functions from ``torch.nn.loss or a custom function defined by the user.
	'loss_func': None, # [object,object] (required)

	# A hybrid QNN model to
	# be trained.
	'model': None, # object (required)

	# A PyTorch optimizer to tune
	# the model parameters.
	'optimizer': None, # object (required)

	# A PyTorch data loader object
	# containing the training dataset.
	'train_data': None, # object (required)
}

service = QiskitRuntimeService(
	channel='ibm_quantum'
)

job = service.run(
	program_id='torch-train',
	options=options,
	inputs=runtime_inputs,
	instance='ibm-q/open/main'
)

# Job id
print(job.job_id)
# See job status
print(job.status())

# Get results
result = job.result()