View Single Post
  #2   Spotlight this post!  
Unread 28-03-2015, 20:42
Ether's Avatar
Ether Ether is offline
systems engineer (retired)
no team
 
Join Date: Nov 2009
Rookie Year: 1969
Location: US
Posts: 8,086
Ether has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond reputeEther has a reputation beyond repute
Re: Analysis of team ranking

Quote:
Originally Posted by Ether View Post

1) compute N = AT∙A

2) compute d = AT∙b

3) compute teamscores = scipy.linalg.cho_solve(scipy.linalg.cho_factor(N), d)

... where A is np.array(matchdata) and b is np.array(matchscores)

The computation time should be reduced from 68 seconds to about 2 seconds or less.
I'd run this test myself but I have Python2.7.5 installed and your Python3 code crashes when I try to run it. Not being very fluent in Python, I'm not in a good position to try to port it.

Based on some testing I did here, I'm fairly confident that your computation time can be dramatically reduced by making the small changes shown above.



Last edited by Ether : 28-03-2015 at 22:13.