why does timeit module take longer time than the code produces output quickly in python?

Question

here is the code to calculate fibonacci

import timeit
counter=0
def fibhelper(n):
  global counter
  counter+=1
  if n==0 :
     return 0
  elif n==1:
     return 1
  else:
     return fibhelper(n-1)+fibhelper(n-2)

print fibhelper(20)
print "Total function calls-- ",counter
t1=timeit.Timer('fibhelper(20)',"from __main__ import fibhelper")
y=t1.timeit()
print "normal method in secs: ",y

output is:

6765
Total function calls-- 21891

which comes out immediately, but it is still calculating y. why is this? when the function is evaluated quickly, why does timeit of that function takes longer?

Your output doesn't match your `print` statements. Why are you using a global `counter` here? Note that `timeit` runs your function 1 million times, not just once. — Martijn Pieters, Nov 12 '13 at 18:24
I asked a similar question here, HTH http://stackoverflow.com/q/19473800/1860929 — Anshul Goyal, Nov 12 '13 at 18:32

score 3 · Accepted Answer · answered Nov 12 '13 at 18:29

3

The default parameters of timeit include: number=1000000.

Quoting the documentation of timeit:

... run its timeit() method with number executions.

Therefore, it is expected to take 1000000 times longer.

answered Nov 12 '13 at 18:29

Has QUIT--Anony-Mousse

76,138
12
138
194

why does timeit module take longer time than the code produces output quickly in python?

1 Answers1