Re: [math] refactoring least squares

2014-02-24 Thread Gilles
On Mon, 24 Feb 2014 14:41:34 -0500, Evan Ward wrote: On 02/24/2014 01:23 PM, Gilles wrote: On Mon, 24 Feb 2014 11:49:26 -0500, Evan Ward wrote: One way to improve performance would be to provide pre-allocated space for the Jacobian and reuse it for each evaluation. Do you have actual data to

[continuum] BUILD FAILURE: Apache Commons Compress - Apache Commons (Group (shared) Maven 3 Build Definition (Java 1.6))

2014-02-24 Thread Apache Continuum
Online report : https://continuum-ci.apache.org/continuum/buildResult.action?buildId=28220&projectId=64 Build statistics: State: Failed Previous State: Ok Started at: Mon 24 Feb 2014 21:20:08 + Finished at: Mon 24 Feb 2014 21:20:40 + Total time: 32s Build Trigger: Schedule B

Re: [math] refactoring least squares

2014-02-24 Thread Evan Ward
On 02/24/2014 02:18 PM, Ted Dunning wrote: > > On Mon, Feb 24, 2014 at 10:23 AM, Gilles > wrote: > > One way to improve performance would be to provide > pre-allocated space > for the Jacobian and reuse it for each evaluation. > > >

Re: [VOTE] Move Commons Attributes to dormant

2014-02-24 Thread Benedikt Ritter
This vote passes with +1 votes from: Matt Benson, Emmanuel Bourg, Stefan Bodewig, Jörg Schaible, Benedikt Ritter all votes are binding, no other votes have been casted. I'll take care of moving Attributes to dormant at the end of the week. Benedikt 2014-02-20 19:48 GMT+01:00 Benedikt Ritter :

Re: [math] refactoring least squares

2014-02-24 Thread Evan Ward
On 02/24/2014 01:23 PM, Gilles wrote: > On Mon, 24 Feb 2014 11:49:26 -0500, Evan Ward wrote: >> One way to improve performance would be to provide pre-allocated space >> for the Jacobian and reuse it for each evaluation. > > Do you have actual data to back this statement? I did some tests with th

Re: [math] refactoring least squares

2014-02-24 Thread Ted Dunning
On Mon, Feb 24, 2014 at 10:23 AM, Gilles wrote: > One way to improve performance would be to provide pre-allocated space >> for the Jacobian and reuse it for each evaluation. >> > > Do you have actual data to back this statement? > > > The >> LeastSquaresProblem interface would then be: >> >> voi

Re: [math] refactoring least squares

2014-02-24 Thread Gilles
On Mon, 24 Feb 2014 11:49:26 -0500, Evan Ward wrote: I've looked into improving performance further, but it seems any further improvements will need big API changes for memory management. Currently using Gauss-Newton with Cholesky (or LU) requires 4 matrix allocations _each_ evaluation. The obj

Re: [math] refactoring least squares

2014-02-24 Thread Gilles
On Mon, 24 Feb 2014 18:09:26 +0100, Luc Maisonobe wrote: Hi Evan, Le 24/02/2014 17:49, Evan Ward a écrit : I've looked into improving performance further, but it seems any further improvements will need big API changes for memory management. Currently using Gauss-Newton with Cholesky (or LU)

[VOTE][RESULT] Release Apache Commons Pool 2.2 RC2 as 2.2

2014-02-24 Thread Mark Thomas
The following votes were cast: Binding: +1: markt, ggregory, psteitz, oheger, joehni, bodewig No other votes were cast. This vote therefore passes. Mark - To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additiona

Re: [math] refactoring least squares

2014-02-24 Thread Luc Maisonobe
Hi Evan, Le 24/02/2014 17:49, Evan Ward a écrit : > I've looked into improving performance further, but it seems any further > improvements will need big API changes for memory management. > > Currently using Gauss-Newton with Cholesky (or LU) requires 4 matrix > allocations _each_ evaluation. Th

Re: [math] refactoring least squares

2014-02-24 Thread Evan Ward
I've looked into improving performance further, but it seems any further improvements will need big API changes for memory management. Currently using Gauss-Newton with Cholesky (or LU) requires 4 matrix allocations _each_ evaluation. The objective function initially allocates the Jacobian matrix.