It is worth noting on this thread that the LightGBM package actually
doesn't have a CUDA backend for Windows. It requires an OpenCL SDK as
noted here (https://github.com/Microsoft/LightGBM/issues/1236).
Regards,
Charles
On Tue, Jan 23, 2018 at 11:10 AM, Emmanuel Hamel
wrote:
> Here is a messag
Hi Emmanuel,
As Dirk mentioned, I am involved in making GPU computing as portable and
accessible as possible. Depending on your needs, most functionality may
already be implemented in my gpuR package (
https://cran.r-project.org/package=gpuR). It is an OpenCL backend so it
can be installed essen
help make gpu computing more accessible to R users is my goal.
Charles
On Tue, Jan 9, 2018 at 9:42 AM, Dirk Eddelbuettel wrote:
>
> On 9 January 2018 at 09:25, Charles Determan wrote:
> | tied to NVIDIA. That said, NVIDIA hardware is difficult to beat and
> | creating packages that
Greetings,
I have been exploring GPU implementations for R. Primarily I have been
using OpenCL given it is so portable and everyone can use it without being
tied to NVIDIA. That said, NVIDIA hardware is difficult to beat and
creating packages that leverage are arguably valuable (e.g. gputools,
g
Greetings,
I have run in a problem where a user tries to install my package 'gpuR' and
the compilation is showing the C++ compiler absent resulting in errors like:
I/usr/lib64/microsoft-r/3.3/lib64/R/include -DNDEBUG -I../inst/include
-DCL_HPP_MINIMUM_OPENCL_VERSION=110 -DCL_USE_DEPRECATED_OPENCL
y good way to
> do it is header-only classes that you put in inst/include.
>
> On Mon, Aug 15, 2016 at 6:17 AM Charles Determan
> wrote:
>
>> Greetings,
>>
>> I am trying to integrate some packages of mine so they can share some of
>> the same C++ functions (n
Greetings,
I am trying to integrate some packages of mine so they can share some of
the same C++ functions (not exposed to the R front-end user). Here is my
basic structure of the package I wish to export functions.
Class defined in header (foo.h) complete with all public and private
members.
M
I am trying to help a user debug the installation of one of my packages and
have come across an error I don't understand.
The current process has been as follows:
1. User using Windows OS
2. User trying to install development version from my github
3. User has current Rtools installed
4. User beh
al file HOME/.R/Makevars.win
> (See R-admin 6.3.3). On Windows, "HOME" is usually your Documents
> folder.
>
> Martyn
>
> On Tue, 2015-11-10 at 10:45 -0600, Charles Determan wrote:
> > The std=c++0x is not sufficient for my purposes as it doesn't support
> > t
es
On Tue, Nov 10, 2015 at 10:42 AM, Martyn Plummer wrote:
> On Tue, 2015-11-10 at 10:30 -0600, Dirk Eddelbuettel wrote:
> > On 10 November 2015 at 10:18, Charles Determan wrote:
> > | Okay, is there a way for me to use the newer compiler for building the
> package
> > | on
:16 AM, Dirk Eddelbuettel > wrote:
>
>
> On 10 November 2015 at 10:00, Charles Determan wrote:
>
> | I have installed a more recent compiler and would like to use the -std=c++11
> | flag instead of the -std=c++0x.
> |
> | system('g++ --version')
> | truncated.
15 at 10:00, Charles Determan wrote:
> | I have installed a more recent compiler and would like to use the
> -std=c++11
> | flag instead of the -std=c++0x.
> |
> | system('g++ --version')
> | truncated... 5.2.0
> |
> | I have a Makevars.win file where I initially
I have installed a more recent compiler and would like to use the
-std=c++11 flag instead of the -std=c++0x.
system('g++ --version')
truncated... 5.2.0
I have a Makevars.win file where I initially set the CXX_STD=CXX11 however
this uses the default C++11 flag returned from R CMD config CXX1XSTD (
ixXf A = *ptrA
> return wrap(A)
> }
>
> On Fri, Jun 26, 2015 at 8:32 AM, Dirk Eddelbuettel wrote:
>
>>
>> On 26 June 2015 at 07:43, Charles Determan wrote:
>> | I have tried a few more feeble attempts (explicitly create pointer to
>> pass,
>> | passing
2015 at 07:43, Charles Determan wrote:
> | I have tried a few more feeble attempts (explicitly create pointer to
> pass,
> | passing just the MatrixXf object) at getting the Eigen object to persist
> | between functions without the additional Armadillo wrapper but with no
> |
, Jun 25, 2015 at 10:34 AM, Charles Determan
wrote:
> I have a question similar to a previous question I asked here regarding
> the passing of XPtr objects between functions (
> https://www.mail-archive.com/rcpp-devel@lists.r-forge.r-project.org/msg08389.html).
> Exploring the same to
I have a question similar to a previous question I asked here regarding the
passing of XPtr objects between functions (
https://www.mail-archive.com/rcpp-devel@lists.r-forge.r-project.org/msg08389.html).
Exploring the same topic with the Eigen library I have run in to another
curious situation. As
a MatrixXf object. I
> suppose you can make a copy of the source matrix and cast it to float type
> using the cast() method.
>
>
> Best,
> Yixuan
>
> 2015-06-25 3:31 GMT+08:00 Charles Determan :
>
>> I have been exploring the use of RcppEigen more recently and stumbled
>&g
Thanks Hadley, much more concise :)
On Wed, Jun 24, 2015 at 2:22 PM, Hadley Wickham wrote:
> On Wed, Jun 24, 2015 at 11:08 AM, Charles Determan
> wrote:
> > Many thanks Krzysztof, your suggestion works. I can explicitly create a
> > 'new' arma::mat object and pas
I have been exploring the use of RcppEigen more recently and stumbled upon
an issue when passing R objects. Is there a way to pass an R matrix and
get it in to a mapped Eigen MatrixXf object? I know R doesn't have a
native float type so I am not overly surprised that I have run in to this
issue.
t B = arma::mat( (double *) ptrB->memptr(),
nr,
nc,
false);
B.print("copied matrix");
}
Regards,
Charles
On Wed, Jun 24, 2015 at 10:19 AM, Krzysztof Sakrejda <
krzysztof.sakre...@gmail.com> wrote:
> Hi Charles, comments inline
>
> 2015-
ma::mat( (double *) ptrB->memptr(),
nr,
nc,
false);
B.print("copied matrix");
}
Charles
On Wed, Jun 24, 2015 at 8:32 AM, Dirk Eddelbuettel wrote:
>
> On 24 June 2015 at 08:22, Charles Determan wrote:
> | Thank you John,
> |
> | I
Thank you John,
I am familiar with bigmemory (I am one of the current developers
actually). The project I am working on doesn't need the shared memory
aspect so was intending to avoid the dependency and just leverage the more
familiar and developed Armadillo library. However your response inform
Hi Dirk,
I was actually looking at that very page. However, I am still not sure
what I am missing here. I am not passing any pointers to functions, just
the data object. Like I said, I feel like I must be missing something
obvious but cannot figure it out. I fail to see what is distinguishing
I am exploring the use of some XPtr objects. For example, I can create a
pointer to an armadillo matrix. I can then create a copy of that matrix
from the pointer within the same function. However, when I try to pass a
pointer to a function and create a new copy from said pointer it fails
(return
ges/sig%20twitter.jpeg]
> <https://twitter.com/Nexidia> [image:
> http://host.msgapp.com/Extranet/96621/Signature%20Images/sig%20Youtube.jpeg]
> <https://www.youtube.com/user/NexidiaTV>
>
>
>
> *From:* rcpp-devel-boun...@lists.r-forge.r-project.org [mailto:
> rcpp-d
Sounds good, thanks Dirk.
On Wed, May 27, 2015 at 8:04 AM, Dirk Eddelbuettel wrote:
>
> On 27 May 2015 at 07:09, Charles Determan wrote:
> | This was more of a curiosity question. Armadillo and Eigen are great
> and I
> | will likely use them for most of my purposes. I
t; On 26 May 2015 at 15:32, Charles Determan wrote:
> | I have seen in the archives a discussion of converting a boost ublas
> matrix to
> | Rcpp (http://lists.r-forge.r-project.org/pipermail/rcpp-devel/2011-June/
> | 002402.html) but I didn't see anything about passing an R matr
I have seen in the archives a discussion of converting a boost ublas matrix
to Rcpp (
http://lists.r-forge.r-project.org/pipermail/rcpp-devel/2011-June/002402.html)
but I didn't see anything about passing an R matrix in to a ublas object.
There doesn't appear to be an 'as' method for it (unless I h
I am actually working on a general purpose GPU library for R using Rcpp and
RcppArmadillo but it is still under heavy development. During these very
early stages I have had an 'older' card (AMD Radeon HD 5700 Series) so I
have been working primarily with OpenCL and the clBLAS library (which must
b
This is probably the best place I can think of to ask this question if
perhaps not completely on topic. Some quick context:
I am exploring GPGPU computing and with my 'open-source' mindset I quickly
gravitated to using OpenCL. Thankfully, many wonderful programmers have
already begun creating li
ave tried to read everything I
> could before asking this list.
>
> ___
> Rcpp-devel mailing list
> Rcpp-devel@lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
>
>
>
>
--
Dr. Charles Determan, PhD
Integrated Biosciences
___
Rcpp-devel mailing list
Rcpp-devel@lists.r-forge.r-project.org
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
I think you want to check your DESCRIPTION file. I believe you need to
have the 'LinkingTo:' option to also include BH.
Regards,
Charles
On Wed, Mar 11, 2015 at 6:58 AM, Jordi Molins Coronado <
jordi_mol...@hotmail.com> wrote:
> Hello,
>
> I know this question has been answered before (and I ha
ff of the length function). Also, in my
> opinion, it is obsolete, especially for Rcpp code, since it is for writing
> code that works in R and S (including S+) and Rcpp will never work in S+.
>
> Bill Dunlap
> TIBCO Software
> wdunlap tibco.com
>
> On Wed, Feb 25, 2015
package now compiles without error.
Thanks for your help Dirk, I know this was a long chain back and forth.
Regards,
Charles
On Wed, Feb 25, 2015 at 7:36 AM, Charles Determan Jr
wrote:
> My apologies, I should have just did this from the start. I thought it
> may be something simple. I create
t; http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org
>
--
Dr. Charles Determan, PhD
Integrated Biosciences
___
Rcpp-devel mailing list
Rcpp-devel@lists.r-forge.r-project.org
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
;refblas64longlong.h"
#define INT long long
#else
#include
#include
#define INT int
#endif
#include "bigmemory/BigMatrix.h"
#include
On Tue, Feb 24, 2015 at 3:29 PM, Dirk Eddelbuettel wrote:
>
> On 24 February 2015 at 15:06, Charles Determan Jr wrote:
> | Unfortunately I
015 at 2:51 PM, Dirk Eddelbuettel wrote:
>
> On 24 February 2015 at 14:41, Charles Determan Jr wrote:
> | Thank you for that clarification, I always seem to forget that Rcpp
> packages
> | have many libraries loaded by default. I have trimmed the headers down
> to the
Eddelbuettel wrote:
>
> On 24 February 2015 at 12:30, Charles Determan Jr wrote:
> | Greetings,
> |
> | I have run in to an error when compiling an R package that has the
> following
> | headers in the only cpp file at the moment:
> |
> | #include
> | #inclu
ct something must be
conflicted between the R headers and RcppArmadillo because it works just
fine if I only use Rcpp.h (but I want to use RcppArmadillo). Any thoughts?
Regards,
--
Dr. Charles Determan, PhD
Integrated Biosciences
___
Rcpp-devel mai
40 matches
Mail list logo
