Get Time
Search | Watch Thread  |  My Post History  |  My Watches  |  User Settings
View: Flat (newest first)  | Threaded  | Tree
Not new | Reply
Just a minor nit, but the solution presented (as stated in the article) is not new. It's been around a long time. A better title might be "A Different Approach ... " or (based on the performance results presented in numerous papers) "A Faster Approach ... ".

It may be new to some TopCoders, certainly, and it is an excellent article, but the headline teases something the article does not deliver.
Re: Not new (response to post by radeye) | Reply
+1 for that. I borrowed the title from the research paper mentioned in the reference. I agree that the approach has been around for more than twenty years. But it was new for me when I read it last month.
about initialize_preflow() (response to post by NilayVaish) | Reply
when initialize,for edge(u,v) where u isn't source, Shall we set
cf[u][v] to c[u][v]?
Re: about initialize_preflow() (response to post by GuoGuo) | Reply
Yes. That's correct. Actually the code that I had written had this line in the main function. And I forgot to put it in the initialize function. You can do a memcpy(cf,c,sizeof(int)*n*n);