#include <bits/stdc++.h>

using namespace std;

ifstream f("flux.in");
ofstream g("flux.out");

struct flux
{
    int x,y,c;
}v[5010];

int n,k,col,m,i,j;
double mini=INT_MAX,suma,a[105][105],sol[105];

int main()
{
    f>>n>>m;
    for(i=1; i<=m; i++)
    {
        f>>v[i].x>>v[i].y>>v[i].c;

        a[v[i].x][v[i].x]++;
        a[v[i].y][v[i].y]++;

        a[v[i].x][v[i].y]--;
        a[v[i].y][v[i].x]--;
    }

    for(i=1; i<=n+1; i++)
    {
        a[1][i]=0;
    }
    a[1][1]=1;
    a[n][n+1]=1;



    i=2;j=2;
    while(i<=n && j<=n)
    {
        for(k=i;k<=n;k++)
            if(a[k][j]!=0) break;

        if(k==n+1)
        {
            j++;
            continue;
        }

        if(k!=i) // interschimb liniile
        {
            for(col=j;col<=n+1;col++)
            {
                swap(a[i][col],a[k][col]);
            }
        }


        // de acum incolo, am facut interschimbarea si lucrez cu i, nu cu k

        for(col=j+1;col<=n+1;col++)
            a[i][col]=a[i][col]/a[i][j];

        a[i][j]=1; // abia la final fac a[i][j]=1, pentru ca altfel se modifica pe parcurs

        for(k=i+1;k<=n;k++)
        {
            for(col=j+1;col<=n+1;col++)
                a[k][col]-=a[k][j]*a[i][col];

            a[k][j]=0; // abia acum il fac 0
        }

        i++;
        j++;
    }

    for(i=n;i>=2;i--)
    {
        for(j=2;j<=n+1;j++)
        {
            if(abs(a[i][j]) > 0.0001)
            {
                if(j==m+1) // am ajuns la coada si rezulta ca 0 = a[i][j]
                {
                    g<<"Imposibil";
                    return 0;
                }

                sol[j]=a[i][n+1];
                for(k=j+1;k<=n;k++)
                    sol[j]-=sol[k]*a[i][k]; // ii scad pe cei anteriori

                break; // nu mai avansez cu j-ul odata ce am gasit un coeficient nenul
            }
        }
    }


   // for(i=1;i<=n;i++)g<<sol[i]<<" ";

    for(i=1;i<=m;i++)
    {
        double rez=v[i].c/(abs(sol[v[i].x]-sol[v[i].y]));
        mini=min(mini,rez);
    }

    if(mini==INT_MAX)
    {
        g<<0;
        return 0;
    }

    for(i=1;i<=m;i++)
    {
        if(v[i].x==n || v[i].y==n)
        {
            suma=suma+mini*abs(sol[v[i].x]-sol[v[i].y]);
        }
    }

    g<<fixed<<setprecision(6)<<mini;
    return 0;
}