Hi Kent.
sum_norm_ted is sum over each i, and all j. That is
why it is initialize to zero at the beginning of each
i loop.
The inner loop IS running - which I checked by all the
print statements. The value of "ted" and "mag" ARE
being calculated correctly as checked by the "print"
statement. But when we do
norm_ted = (2 * ted) / mag
becomes zero, and subsequently sum_norm_ted becomes
zero.
--- Kent Johnson <kent37 at tds.net> wrote:
> Sanhita Mallick wrote:
>
> > I am struggling with a simple recurring function
>
> What is a recurring function? Do you mean looping?
> >
> > Here is the script. T(i) are trees (as in graphs).
> > The program operates great, until the step where
> > norm_ted is calculated. Mysteriously norm_ted
> becomes
> > 0, even though "mag" value is calculated
> correctly.
> > Also, strangely, sum_norm_ted also becomes zero.
> >
> > What am I missing?
>
> Do you mean for sum_norm_ted to be the sum over all
> i and j? If so, you
> should initialize it before the start of the outer
> loop.
>
> I don't know why you are getting 0 at the end of the
> loop. It is as if
> sum_norm_ted and norm_ted are initialized but the
> inner loop doesn't
> run. Is this the exact code that you have trouble
> with?
>
> Kent
> >
> > Thanks.
> > -Sanhita
> >
> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > T = [F1, F2, F3]
> > k = len(T)
> > mean_ted = {}
> > for i in range(k):
> > sum_norm_ted = 0.0
> > mag = 0.0
> > norm_ted = 0.0
> > for j in range(k):
> > if (i != j):
> > ted= alg.ted(T[i], T[j])
> > print 'TED ' + str(i) + str(j) + ' is
> ' +
> > str(ted)
> > mag = 2 * (T[i].n + T[j].n) - 1
> > print 'MAG ' + str(i) + str(j) + ' is
> ' +
> > str (mag) + ' and ' + str ( 2 * ted )
> > norm_ted = (2 * ted) / mag
> > print 'norm_ted ' + str(i) + str(j) +
> ' is
> > ' + str (norm_ted)
> > sum_norm_ted = sum_norm_ted + norm_ted
> > print 'sum_norm_ted ' + str(i) +
> str(j) +
> > ' is ' + str (sum_norm_ted)
> > _______________________________________________
> > Tutor maillist - Tutor at python.org
> > http://mail.python.org/mailman/listinfo/tutor
> >
>
>