Re: Matrix operations on large datasets

[Ben Pfaff]

Ed <address@hidden> writes:

> (To avoid polluting the standard thread I copied this part across)
>
>> 2008/5/30 Ben Pfaff <address@hidden>:
>> But I'm not sure why you think that supporting large data sets
>> has to be highly complex.  A lot of it is just Computer Science
>> 101 type stuff; for example, merge sorts for efficiently sorting
>> large data sets (which we already implement).  What do you have
>> in mind?
>
> I guess I was thinking of matrix operations in particular. I've been
> reading the linear regression code as Jason suggested, and the best
> way to implement linear algebra on a large dataset isn't immediately
> clear to me.

You're right.  We do not have any provisions for out-of-core
linear algebra.  Whether this is an important omission, I will
have to defer to others.
-- 
"Mon peu de succès près des femmes est toujours venu de les trop aimer."
--Jean-Jacques Rousseau