swarm-modeling
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Question on partitioning a set of agents.


From: Theodore C. Belding
Subject: Re: Question on partitioning a set of agents.
Date: Sun, 8 Aug 1999 13:20:33 -0400 (EDT)

Sorry --- I misunderstood what you were trying to do and misread
equiprobable partitions to be "equal-sized partitions". If you're trying
to do that (divide the agents into N equal-sized partitions), you can
still use the Shuffler, as in my earlier message. If you're just trying to
divide up the agents into N partitions, with an equal probability for each
partition, and you don't care about keeping partition sizes equal, then
just go through the list and roll an N-sided die (pick a random int from 1
to N) for each object to pick its partition; there's still no need to make
multiple passes through the list. 
-Ted

--
Ted Belding                              address@hidden 
University of Michigan Center for the Study of Complex Systems
Homepage: http://www-personal.umich.edu/~streak/
PGP key:  http://www-personal.umich.edu/~streak/pgp-key.html

On Fri, 6 Aug 1999, Jason Alexander wrote:

> For one of my models I need to randomly partition a set of agents
> (stored as a list) into N disjoint classes, where each class is
> equiprobable.
> 
> Are there any implicit biases in the following procedure that I'm
> unaware of?  (It seems okay to me.)
> 
> Let L denote the list of agents.
> Assume RandomFloat() return a number between 0 and 1.
> 
> for i=1 upto N-1 {
>   loop over L {
>      if RandomFloat < 1/N {
>         remove current agent from L
>         insert current agent into class i
>      }
>   }
> }
> Copy remaining agents from L into class N.
> 
> 
> Cheers,
> 
> Jason
> 
> 
>                   ==================================
>    Swarm-Modelling is for discussion of Simulation and Modelling techniques
>    esp. using Swarm.  For list administration needs (esp. [un]subscribing),
>    please send a message to <address@hidden> with "help" in the
>    body of the message.
>                   ==================================
> 



                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================


reply via email to

[Prev in Thread] Current Thread [Next in Thread]