[Rcpp-devel] [Pre-ANN] RcppArmadillo 0.7.960.1.0

[Dirk Eddelbuettel]

RcppArmadillo 0.7.960.1.0 has been ready for a few days and sitting at CRAN,
due to a seemingly unlucky combination of travel, weekend,, systems outage
and what not. But it has been five full days now , and nobody has told me
what if anything is non-standard with the package so ... I give up now.

So it is as of now also in the drat repo for the RcppCore org, so just see
the README.md at
https://github.com/RcppCore/drat
which has the following example which is applicable:

# first add the repo
drat:::add("RcppCore")
# either install just one or more given packages
install.packages("RcppArmadillo")
# or update already installed packages
update.packages()

You can also add the repo URL by hand to options("repos"), or supply it to
install.packages(), or ... I happen to like drat.

The files NEWS.Rd and ChangeLog have the goods, I make a fuller announcement
if and when it makes it to CRAN.

This release features a lot of excellent Google Summer of Code work by
Binxiang Ni. Again, more about that if and when we get to make a true
release announcement.

Dirk

[Kurt Hornik]

Dirk:

When checking the reverse depends, we're seeing problems with quanteda,
both with the old version of this as well as the new one published last
night. E.g.

Package: quanteda [Old version: 0.9.9-65, New version: 0.99]
Check: examples
New result: ERROR
Running examples in ‘quanteda-Ex.R’ failed

base::assign(".ptime", proc.time(), pos = "CheckExEnv")
### Name: textmodel_wordfish
### Title: wordfish text model
### Aliases: textmodel_wordfish
### ** Examples
textmodel_wordfish(data_dfm_lbgexample, dir = c(1,5))

Error in wordfishcpp_mt(x, as.integer(dir), 1/(priors^2), tol, disp, dispersion_floor, :
dfmSparse is not supported.
Calls: textmodel_wordfish -> textmodel_wordfish.dfm -> wordfishcpp_mt -> .Call
Execution halted

and in fact all problems seem to have the "dfmSparse is not supported."
errors. Are you seeing these too? Should we contact the quanteda
maintainer about this?

Best
-k

RcppArmadillo 0.7.960.1.0 has been ready for a few days and sitting at CRAN,
due to a seemingly unlucky combination of travel, weekend,, systems outage
and what not. But it has been five full days now , and nobody has told me
what if anything is non-standard with the package so ... I give up now.
So it is as of now also in the drat repo for the RcppCore org, so just see
the README.md at
https://github.com/RcppCore/drat
# first add the repo
drat:::add("RcppCore")
# either install just one or more given packages
install.packages("RcppArmadillo")
# or update already installed packages
update.packages()
You can also add the repo URL by hand to options("repos"), or supply it to
install.packages(), or ... I happen to like drat.
The files NEWS.Rd and ChangeLog have the goods, I make a fuller announcement
if and when it makes it to CRAN.
This release features a lot of excellent Google Summer of Code work by
Binxiang Ni. Again, more about that if and when we get to make a true
release announcement.
Dirk
--

[Dirk Eddelbuettel]

Hi Kurt,

Much appreciate the update.

On 16 August 2017 at 17:23, Kurt Hornik wrote:
| >>>>> Dirk Eddelbuettel writes:
|
| Dirk:
|
| When checking the reverse depends, we're seeing problems with quanteda,
| both with the old version of this as well as the new one published last
| night. E.g.
|
| Package: quanteda [Old version: 0.9.9-65, New version: 0.99]
| Check: examples
| New result: ERROR
| Running examples in ‘quanteda-Ex.R’ failed
| The error most likely occurred in:
|
| > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
| > ### Name: textmodel_wordfish
| > ### Title: wordfish text model
| > ### Aliases: textmodel_wordfish
| >
| > ### ** Examples
| >
| > textmodel_wordfish(data_dfm_lbgexample, dir = c(1,5))
| Error in wordfishcpp_mt(x, as.integer(dir), 1/(priors^2), tol, disp, dispersion_floor, :
| dfmSparse is not supported.
| Calls: textmodel_wordfish -> textmodel_wordfish.dfm -> wordfishcpp_mt -> .Call
| Execution halted
|
| and in fact all problems seem to have the "dfmSparse is not supported."
| errors. Are you seeing these too? Should we contact the quanteda
| maintainer about this?

I have been seeing _fairly consistent_ failures for quanteda with both Rcpp
and RcppArmadillo for quite some time. You can see more details at the usual
rcpp-logs repo.

As such, I have ignored its most recent failure with RcppArmadillo
0.7.960.1.0. The last (of several) reverse depends logs preceding the upload
has its summary here:
https://github.com/RcppCore/rcpp-logs/blob/master/results/rcpparmadillo/RcppArmadillo-Summary-20170810.txt
As you can see: 372 passed, 3 failed and 2 were skipped by my choice.

On occassion, I dive into packages that fail, ie Rob's "forecast" apparently
only failed for me but not elsewhere. As it was reproducible, Rob and I
agreed on a change he made. So it no longer fails for me either.

Similarly, preceding the currently-pending upload of a new BH I have reached
out to two packages one of which was already updated, the other is pending.
But I don't have the bandwidth to chase all failures for all packages, and
some are too persistent. And that is why I uploaded RcppArmadillo last Friday.

Dirk

| Best
| -k
|
|
| > RcppArmadillo 0.7.960.1.0 has been ready for a few days and sitting at CRAN,
| > due to a seemingly unlucky combination of travel, weekend,, systems outage
| > and what not. But it has been five full days now , and nobody has told me
| > what if anything is non-standard with the package so ... I give up now.
|
| > So it is as of now also in the drat repo for the RcppCore org, so just see
| > the README.md at
| > https://github.com/RcppCore/drat
| > which has the following example which is applicable:
|
| > # first add the repo
| > drat:::add("RcppCore")
| > # either install just one or more given packages
| > install.packages("RcppArmadillo")
| > # or update already installed packages
| > update.packages()
|
| > You can also add the repo URL by hand to options("repos"), or supply it to
| > install.packages(), or ... I happen to like drat.
|
| > The files NEWS.Rd and ChangeLog have the goods, I make a fuller announcement
| > if and when it makes it to CRAN.
|
| > This release features a lot of excellent Google Summer of Code work by
| > Binxiang Ni. Again, more about that if and when we get to make a true
| > release announcement.
|
| > Dirk
|
| > --
| > http://dirk.eddelbuettel.com | @eddelbuettel | ***@debian.org
|

[Kurt Hornik]

Thanks. So I guess we should publish the new RcppArmadillo and ask the
quanteda maintainer to fix as necessary?

Best
-k

Post by Dirk Eddelbuettel
Hi Kurt,
Much appreciate the update.
|
|
| When checking the reverse depends, we're seeing problems with quanteda,
| both with the old version of this as well as the new one published last
| night. E.g.
|
| Package: quanteda [Old version: 0.9.9-65, New version: 0.99]
| Check: examples
| New result: ERROR
| Running examples in ‘quanteda-Ex.R’ failed
|
| > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
| > ### Name: textmodel_wordfish
| > ### Title: wordfish text model
| > ### Aliases: textmodel_wordfish
| >
| > ### ** Examples
| >
| > textmodel_wordfish(data_dfm_lbgexample, dir = c(1,5))
| dfmSparse is not supported.
| Calls: textmodel_wordfish -> textmodel_wordfish.dfm -> wordfishcpp_mt -> .Call
| Execution halted
|
| and in fact all problems seem to have the "dfmSparse is not supported."
| errors. Are you seeing these too? Should we contact the quanteda
| maintainer about this?
I have been seeing _fairly consistent_ failures for quanteda with both Rcpp
and RcppArmadillo for quite some time. You can see more details at the usual
rcpp-logs repo.
As such, I have ignored its most recent failure with RcppArmadillo
0.7.960.1.0. The last (of several) reverse depends logs preceding the upload
https://github.com/RcppCore/rcpp-logs/blob/master/results/rcpparmadillo/RcppArmadillo-Summary-20170810.txt
As you can see: 372 passed, 3 failed and 2 were skipped by my choice.
On occassion, I dive into packages that fail, ie Rob's "forecast" apparently
only failed for me but not elsewhere. As it was reproducible, Rob and I
agreed on a change he made. So it no longer fails for me either.
Similarly, preceding the currently-pending upload of a new BH I have reached
out to two packages one of which was already updated, the other is pending.
But I don't have the bandwidth to chase all failures for all packages, and
some are too persistent. And that is why I uploaded RcppArmadillo last Friday.
Dirk
| Best
| -k
|
|
| > RcppArmadillo 0.7.960.1.0 has been ready for a few days and sitting at CRAN,
| > due to a seemingly unlucky combination of travel, weekend,, systems outage
| > and what not. But it has been five full days now , and nobody has told me
| > what if anything is non-standard with the package so ... I give up now.
|
| > So it is as of now also in the drat repo for the RcppCore org, so just see
| > the README.md at
| > https://github.com/RcppCore/drat
|
| > # first add the repo
| > drat:::add("RcppCore")
| > # either install just one or more given packages
| > install.packages("RcppArmadillo")
| > # or update already installed packages
| > update.packages()
|
| > You can also add the repo URL by hand to options("repos"), or supply it to
| > install.packages(), or ... I happen to like drat.
|
| > The files NEWS.Rd and ChangeLog have the goods, I make a fuller announcement
| > if and when it makes it to CRAN.
|
| > This release features a lot of excellent Google Summer of Code work by
| > Binxiang Ni. Again, more about that if and when we get to make a true
| > release announcement.
|
| > Dirk
|
| > --
|
--

[Dirk Eddelbuettel]

On 16 August 2017 at 18:11, Kurt Hornik wrote:
| >>>>> Dirk Eddelbuettel writes:
|
| Thanks. So I guess we should publish the new RcppArmadillo and ask the
| quanteda maintainer to fix as necessary?

And I'll help. I have not interacted much with Ken, but we have mutual
acquaintances and I am sure we can sort this out "pronto".

And I looked some more, and it seems "weird" as it fails more with
RcppArmadillo than with Rcpp (see below). But I am sure we will get this
squared away.

Much appreciate the feedback.

Cheers, Dirk

~/git/rcpp-logs/results/rcpparmadillo(master)$ grep quanteda result-20170*txt
result-20170206-223442.txt:"228","quanteda",0
result-20170411-202011.txt:"250","quanteda","ok"
result-20170503-184823.txt:"251","quanteda","ok"
result-20170504-125802.txt:"251","quanteda","ok"
result-20170516-173035.txt:"253","quanteda","ok"
result-20170523-182911.txt:"253","quanteda","failed"
result-20170525-022447.txt:"253","quanteda","failed"
result-20170531-133609.txt:"253","quanteda","ok"
result-20170619-161821.txt:"258","quanteda","failed"
result-20170801-235421.txt:"271","quanteda","failed"
result-20170803-183132.txt:"272","quanteda","failed"
result-20170808-173708.txt:"272","quanteda","failed"
result-20170810-164134.txt:"272","quanteda","failed"
~/git/rcpp-logs/results/rcpparmadillo(master)$ cd ../
~/git/rcpp-logs/results(master)$ grep quanteda result-20170*txt
result-20170103-192706.txt:"592","quanteda",1
result-20170104-181933.txt:"590","quanteda",1
result-20170105-175736.txt:"590","quanteda",1
result-20170106-183041.txt:"590","quanteda",1
result-20170110-150305.txt:"590","quanteda",0
result-20170216-192443.txt:"623","quanteda",0
result-20170303-041605.txt:"634","quanteda",0
result-20170313-235138.txt:"643","quanteda",0
result-20170316-080319.txt:"645","quanteda",0
result-20170330-114205.txt:"651","quanteda","ok"
result-20170414-061229.txt:"661","quanteda","ok"
result-20170414-232420.txt:"663","quanteda","ok"
result-20170415-224120.txt:"664","quanteda","ok"
result-20170418-002743.txt:"660","quanteda","ok"
result-20170419-211655.txt:"664","quanteda","ok"
result-20170507-012101.txt:"677","quanteda","ok"
result-20170508-000918.txt:"677","quanteda","ok"
result-20170509-044332.txt:"677","quanteda","ok"
result-20170519-031106.txt:"686","quanteda","ok"
result-20170615-072413.txt:"703","quanteda","ok"
result-20170619-004031.txt:"710","quanteda","ok"
result-20170630-045018.txt:"720","quanteda","failed"
result-20170706-042937.txt:"724","quanteda","ok"
result-20170709-132757.txt:"727","quanteda","ok"
result-20170711-085311.txt:"729","quanteda","ok"
~/git/rcpp-logs/results(master)$

[Kurt Hornik]

Post by Dirk Eddelbuettel
|
| Thanks. So I guess we should publish the new RcppArmadillo and ask the
| quanteda maintainer to fix as necessary?
And I'll help. I have not interacted much with Ken, but we have mutual
acquaintances and I am sure we can sort this out "pronto".
And I looked some more, and it seems "weird" as it fails more with
RcppArmadillo than with Rcpp (see below). But I am sure we will get this
squared away.

Ok. New RcppArmadillo on CRAN now.

Should we write to Ken?

-k

Post by Dirk Eddelbuettel
Much appreciate the feedback.
Cheers, Dirk
~/git/rcpp-logs/results/rcpparmadillo(master)$ grep quanteda result-20170*txt
result-20170206-223442.txt:"228","quanteda",0
result-20170411-202011.txt:"250","quanteda","ok"
result-20170503-184823.txt:"251","quanteda","ok"
result-20170504-125802.txt:"251","quanteda","ok"
result-20170516-173035.txt:"253","quanteda","ok"
result-20170523-182911.txt:"253","quanteda","failed"
result-20170525-022447.txt:"253","quanteda","failed"
result-20170531-133609.txt:"253","quanteda","ok"
result-20170619-161821.txt:"258","quanteda","failed"
result-20170801-235421.txt:"271","quanteda","failed"
result-20170803-183132.txt:"272","quanteda","failed"
result-20170808-173708.txt:"272","quanteda","failed"
result-20170810-164134.txt:"272","quanteda","failed"
~/git/rcpp-logs/results/rcpparmadillo(master)$ cd ../
~/git/rcpp-logs/results(master)$ grep quanteda result-20170*txt
result-20170103-192706.txt:"592","quanteda",1
result-20170104-181933.txt:"590","quanteda",1
result-20170105-175736.txt:"590","quanteda",1
result-20170106-183041.txt:"590","quanteda",1
result-20170110-150305.txt:"590","quanteda",0
result-20170216-192443.txt:"623","quanteda",0
result-20170303-041605.txt:"634","quanteda",0
result-20170313-235138.txt:"643","quanteda",0
result-20170316-080319.txt:"645","quanteda",0
result-20170330-114205.txt:"651","quanteda","ok"
result-20170414-061229.txt:"661","quanteda","ok"
result-20170414-232420.txt:"663","quanteda","ok"
result-20170415-224120.txt:"664","quanteda","ok"
result-20170418-002743.txt:"660","quanteda","ok"
result-20170419-211655.txt:"664","quanteda","ok"
result-20170507-012101.txt:"677","quanteda","ok"
result-20170508-000918.txt:"677","quanteda","ok"
result-20170509-044332.txt:"677","quanteda","ok"
result-20170519-031106.txt:"686","quanteda","ok"
result-20170615-072413.txt:"703","quanteda","ok"
result-20170619-004031.txt:"710","quanteda","ok"
result-20170630-045018.txt:"720","quanteda","failed"
result-20170706-042937.txt:"724","quanteda","ok"
result-20170709-132757.txt:"727","quanteda","ok"
result-20170711-085311.txt:"729","quanteda","ok"
~/git/rcpp-logs/results(master)$
--

[Dirk Eddelbuettel]

On 17 August 2017 at 14:42, Kurt Hornik wrote:
| >>>>> Dirk Eddelbuettel writes:
|
| > On 16 August 2017 at 18:11, Kurt Hornik wrote:
| > | >>>>> Dirk Eddelbuettel writes:
| > |
| > | Thanks. So I guess we should publish the new RcppArmadillo and ask the
| > | quanteda maintainer to fix as necessary?
|
| > And I'll help. I have not interacted much with Ken, but we have mutual
| > acquaintances and I am sure we can sort this out "pronto".
|
| > And I looked some more, and it seems "weird" as it fails more with
| > RcppArmadillo than with Rcpp (see below). But I am sure we will get this
| > squared away.
|
| Ok. New RcppArmadillo on CRAN now.

Thank you.

| Should we write to Ken?

I will do that (and CC CRAN) once I get a chance to look more closely at the
issue, and possibly remedies.

Dirk


| -k
|
| > Much appreciate the feedback.
|
| > Cheers, Dirk
|
| > ~/git/rcpp-logs/results/rcpparmadillo(master)$ grep quanteda result-20170*txt
| > result-20170206-223442.txt:"228","quanteda",0
| > result-20170411-202011.txt:"250","quanteda","ok"
| > result-20170503-184823.txt:"251","quanteda","ok"
| > result-20170504-125802.txt:"251","quanteda","ok"
| > result-20170516-173035.txt:"253","quanteda","ok"
| > result-20170523-182911.txt:"253","quanteda","failed"
| > result-20170525-022447.txt:"253","quanteda","failed"
| > result-20170531-133609.txt:"253","quanteda","ok"
| > result-20170619-161821.txt:"258","quanteda","failed"
| > result-20170801-235421.txt:"271","quanteda","failed"
| > result-20170803-183132.txt:"272","quanteda","failed"
| > result-20170808-173708.txt:"272","quanteda","failed"
| > result-20170810-164134.txt:"272","quanteda","failed"
| > ~/git/rcpp-logs/results/rcpparmadillo(master)$ cd ../
| > ~/git/rcpp-logs/results(master)$ grep quanteda result-20170*txt
| > result-20170103-192706.txt:"592","quanteda",1
| > result-20170104-181933.txt:"590","quanteda",1
| > result-20170105-175736.txt:"590","quanteda",1
| > result-20170106-183041.txt:"590","quanteda",1
| > result-20170110-150305.txt:"590","quanteda",0
| > result-20170216-192443.txt:"623","quanteda",0
| > result-20170303-041605.txt:"634","quanteda",0
| > result-20170313-235138.txt:"643","quanteda",0
| > result-20170316-080319.txt:"645","quanteda",0
| > result-20170330-114205.txt:"651","quanteda","ok"
| > result-20170414-061229.txt:"661","quanteda","ok"
| > result-20170414-232420.txt:"663","quanteda","ok"
| > result-20170415-224120.txt:"664","quanteda","ok"
| > result-20170418-002743.txt:"660","quanteda","ok"
| > result-20170419-211655.txt:"664","quanteda","ok"
| > result-20170507-012101.txt:"677","quanteda","ok"
| > result-20170508-000918.txt:"677","quanteda","ok"
| > result-20170509-044332.txt:"677","quanteda","ok"
| > result-20170519-031106.txt:"686","quanteda","ok"
| > result-20170615-072413.txt:"703","quanteda","ok"
| > result-20170619-004031.txt:"710","quanteda","ok"
| > result-20170630-045018.txt:"720","quanteda","failed"
| > result-20170706-042937.txt:"724","quanteda","ok"
| > result-20170709-132757.txt:"727","quanteda","ok"
| > result-20170711-085311.txt:"729","quanteda","ok"
| > ~/git/rcpp-logs/results(master)$
|
| > --
| > http://dirk.eddelbuettel.com | @eddelbuettel | ***@debian.org

[Dirk Eddelbuettel]

A new RcppArmadillo release is now on CRAN after quite some testing on both
our end, and at CRAN. It contains an excellent Google Summer of Code
contribution by Binxiang Ni which greatly expands sparse matrix support, as
well as a few other additions. See the included blog post below, the full
version with hyperlinks is at

http://dirk.eddelbuettel.com/blog/2017/08/17#rcpparmadillo_0.7.960.1.0

Thanks, Dirk


Thu, 17 Aug 2017

RcppArmadillo 0.7.960.1.0

armadillo image

The bi-monthly RcppArmadillo release is out with a new version 0.7.960.1.0
which is now on CRAN, and will get to Debian in due course.

And it is a big one. Lots of nice upstream changes from Armadillo, and
lots of work on our end as the Google Summer of Code project by Binxiang
Ni, plus a few smaller enhancements -- see below for details.

Armadillo is a powerful and expressive C++ template library for linear
algebra aiming towards a good balance between speed and ease of use with a
syntax deliberately close to a Matlab. RcppArmadillo integrates this
library with the R environment and language--and is widely used by
(currently) 379 other packages on CRAN---an increase of 49 since the last
CRAN release in June!

Changes in this release relative to the previous CRAN release are as
follows:

Changes in RcppArmadillo version 0.7.960.1.0 (2017-08-11)

* Upgraded to Armadillo release 7.960.1 (Northern Banana Republic
Deluxe)

* faster randn() when using OpenMP (NB: usually omitted when used
fromR)

* faster gmm_diag class, for Gaussian mixture models with
diagonal covariance matrices

* added .sum_log_p() to the gmm_diag class

* added gmm_full class, for Gaussian mixture models with full
covariance matrices

* expanded .each_slice() to optionally use OpenMP for
multi-threaded execution

* Upgraded to Armadillo release 7.950.0 (Northern Banana Republic)

* expanded accu() and sum() to use OpenMP for processing
expressions with computationally expensive element-wise
functions

* expanded trimatu() and trimatl() to allow specification of the
diagonal which delineates the boundary of the triangular part

* Enhanced support for sparse matrices (Binxiang Ni as part of Google
Summer of Code 2017)

* Add support for dtCMatrix and dsCMatrix (#135)

* Add conversion and unit tests for dgT, dtT and dsTMatrix (#136)

* Add conversion and unit tests for dgR, dtR and dsRMatrix (#139)

* Add conversion and unit tests for pMatrix and ddiMatrix (#140)

* Rewrite conversion for dgT, dtT and dsTMatrix, and add
file-based tests (#142)

* Add conversion and unit tests for indMatrix (#144)

* Rewrite conversion for ddiMatrix (#145)

* Add a warning message for matrices that cannot be converted
(#147)

* Add new vignette for sparse matrix support (#152; Dirk in #153)

* Add support for sparse matrix conversion from Python SciPy
(#158 addressing #141)

* Optional return of row or column vectors in collapsed form if
appropriate #define is set (Serguei Sokol in #151 and #154)

* Correct speye() for non-symmetric cases (Qiang Kou in #150 closing
#149).

* Ensure tests using Scientific Python and reticulate are properly
conditioned on the packages being present.

* Added .aspell/ directory with small local directory now supported by
R-devel.

Courtesy of CRANberries, there is a diffstat report. More detailed
information is on the RcppArmadillo page. Questions, comments etc should
go to the rcpp-devel mailing list off the R-Forge page.

This post by Dirk Eddelbuettel originated on his Thinking inside the box
blog. Please report excessive re-aggregation in third-party for-profit
settings.

/code/rcpp | permanent link