--- Phil Steitz <[EMAIL PROTECTED]> wrote:
> Al Chou wrote:
> > --- Phil Steitz <[EMAIL PROTECTED]> wrote:
> >>I think that is OK, since if you look at the definition of S earlier in
> >>the paper, S is not the variance, it is the sum of the squared
> >>deviations from the mean. This should be a
Al Chou wrote:
--- Phil Steitz <[EMAIL PROTECTED]> wrote:
Al Chou wrote:
Date: Wed, 04 Jun 2003 21:05:14 -0700
From: Phil Steitz <[EMAIL PROTECTED]>
Subject: [math] more improvement to storage free mean, variance computation
Check out procedure sum.2 and var.2 in
http://www.stanfor
--- "Mark R. Diggory" <[EMAIL PROTECTED]> wrote:
> Think I'll bring this back to the list now.
>
> Interesting they actually recommend the two pass for when you can store
> all the values and "Wests" for when storage isn't required/available.
> Maybe we whould evaluate "Wests" for univariateImpl
--- Phil Steitz <[EMAIL PROTECTED]> wrote:
> Al Chou wrote:
> >>Date: Wed, 04 Jun 2003 21:05:14 -0700
> >>From: Phil Steitz <[EMAIL PROTECTED]>
> >>Subject: [math] more improvement to storage free mean, variance computation
> >>
>
Al Chou wrote:
Date: Wed, 04 Jun 2003 21:05:14 -0700
From: Phil Steitz <[EMAIL PROTECTED]>
Subject: [math] more improvement to storage free mean, variance computation
Check out procedure sum.2 and var.2 in
http://www.stanford.edu/~glynn/PDF/0208.pdf
The first looks like Brent's sugge
Think I'll bring this back to the list now.
Interesting they actually recommend the two pass for when you can store
all the values and "Wests" for when storage isn't required/available.
Maybe we whould evaluate "Wests" for univariateImpl and the twopass for
the StoreUnivariate?
-Mark
Al Chou
--- "Mark R. Diggory" <[EMAIL PROTECTED]> wrote:
> Al Chou wrote:
> > After implementing var.2 from the Stanford paper in UnivariateImpl and
> > scratching my head for some time over why the variance calculation failed
> its
> > JUnit test case, I realized there's a flaw in var.2 that I can't under
Al Chou wrote:
After implementing var.2 from the Stanford paper in UnivariateImpl and
scratching my head for some time over why the variance calculation failed its
JUnit test case, I realized there's a flaw in var.2 that I can't understand no
one talks about. To update the variance (called S in
>Date: Wed, 04 Jun 2003 21:05:14 -0700
>From: Phil Steitz <[EMAIL PROTECTED]>
>Subject: [math] more improvement to storage free mean, variance computation
>
>Check out procedure sum.2 and var.2 in
>
>http://www.stanford.edu/~glynn/PDF/0208.pdf
>
>The first lo
Check out procedure sum.2 and var.2 in
http://www.stanford.edu/~glynn/PDF/0208.pdf
The first looks like Brent's suggestion for a corrected mean
computation, with no memory required. The additional computational cost
that I complained about is docuemented to be 3x the flops cost of the
direct
10 matches
Mail list logo
