On 10/17/07, Abandoned <[EMAIL PROTECTED]> wrote: > Very very thanks everbody.. > > These are some method.. > Now the fastest method is second.. > > ==== 1 === > def sortt(d): > items=d.items() > backitems=[ [v[1],v[0]] for v in items] > backitems.sort() > #boyut=len(backitems) > #backitems=backitems[boyut-500:] > a=[ backitems[i][1] for i in range(0,len(backitems))] > a.reverse() > return a > > ==== 2 ===== > import operator > def sortt(d): > backitems=d.items() > boyut=len(backitems) > backitems=backitems[boyut-500:] > backitems=sorted(backitems, key=operator.itemgetter(1)) > a=[ backitems[i][0] for i in range(0,len(backitems))] > a.reverse() > return a > > ==== 3 ===== > def sortt(d): > backitems=d.items() > backitems.sort(lambda x,y:cmp(x[1],y[1])) > backitems=sorted(backitems, key=operator.itemgetter(1)) > a=[ backitems[i][0] for i in range(0,len(backitems))] > a.reverse() > return a > > ====== 4 ======= > import heapq > > def sortt(d): > backitems=d.items() > backitems=heapq.nlargest(1000, backitems, operator.itemgetter(1)) > a=[ backitems[i][0] for i in range(0,len(backitems))] > a.reverse() > return a >
Btw, there are specialized algorithms called "Selection Algorithms" for finding k largest items in a collection. http://en.wikipedia.org/wiki/Selection_algorithm Cheers, -- -- Amit Khemka -- http://mail.python.org/mailman/listinfo/python-list
