Potential Problem with latest R Studio IDE

[thsavage]

For bugs or installation issues, please provide the following information. The more information you provide, the more likely people will be able to help you.

Environment info

Operating System: Windows

Compiler:

Package used (Python/R/Scala/Julia): R

MXNet version: 0.7

Or if installed from source:

MXNet commit hash (git rev-parse HEAD):

If you are using python package, please provide

Python version and distribution:

If you are using R package, please provide

R sessionInfo(): R version 3.3.2 (2016-10-31) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 7 x64 (build 7601) Service Pack 1

locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252 [4] LC_NUMERIC=C LC_TIME=English_United States.1252

attached base packages: [1] stats graphics grDevices utils datasets methods base

other attached packages: [1] mlbench_2.1-1 deepnet_0.2 neuralnet_1.33

loaded via a namespace (and not attached): [1] igraph_1.0.1 Rcpp_0.12.9 rstudioapi_0.6 magrittr_1.5 munsell_0.4.3 colorspace_1.3-2
[7] R6_2.2.0 brew_1.0-6 stringr_1.1.0 plyr_1.8.4 dplyr_0.5.0 visNetwork_1.0.3
[13] Rook_1.1-1 tools_3.3.2 grid_3.3.2 gtable_0.2.0 DBI_0.5-1 influenceR_0.1.0
[19] DiagrammeR_0.9.0 htmltools_0.3.5 lazyeval_0.2.0 digest_0.6.12 assertthat_0.1 tibble_1.2
[25] gridExtra_2.2.1 RColorBrewer_1.1-2 ggplot2_2.2.1 htmlwidgets_0.8 viridis_0.3.4 rgexf_0.15.3
[31] stringi_1.1.2 scales_0.4.1 XML_3.98-1.5 jsonlite_1.2

Error Message:

Loading required package: mxnet Error : object ‘combine_edges’ is not exported by 'namespace:DiagrammeR'

Minimum reproducible example

if you are using your own code, please provide a short script that reproduces the error.

Steps to reproduce

or if you are running standard examples, please provide the commands you have run that lead to the error.

1. 2. 3.

What have you tried to solve it?

1. Tried to run on both Mac and Windows, so it's not the operating system, but could be revised R Studio IDE.

[thirdwing]

This is nothing with Rstudio.

This is really because the Windows lib is old.

Sorry that I have no access to any Windows machine now.

[thsavage]

I've tried with both Windows and Mac. The issue is that you haven't compiled under the most recent version of the R engine. No worries. It was a convenient tool for teaching. Lots of other tools.

Best of luck as you expand.

On Thu, Feb 2, 2017 at 7:50 PM, Qiang Kou (KK) notifications@github.com wrote:

This is nothing with Rstudio.

This is really because the Windows lib is old.

Sorry that I have no access to any Windows machine now.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/dmlc/mxnet/issues/4861#issuecomment-277133073, or mute the thread https://github.com/notifications/unsubscribe-auth/AGIm2Hptj2_N75DTaboqARr-XTQeLhOpks5rYnnTgaJpZM4L1v9y .

-- Tim Savage 917-439-3175

[thirdwing]

The prebuilt pkgs for Windows and OSX are version 0.7.

We have fixed the error introduced by DiagrammeR in the version of 0.9.

Personally I don't have access to machines to update the prebuilt pkgs.

[felixcheung]

I should be able to help, manually, if that would be useful. We could address this in the next release by looking into AppVeyor for Windows build - have other projects that use that

[thirdwing]

Hi, @felixcheung

Thank you for this. I believe you know Windows much better than me.

Can you use other toolchain rather than VS on AppVeyor ?

[jeremiedb]

If it can be useful for some R users in the meantime, I've put compiled binaries that should work with R 3.3 on windows: install.packages("https://github.com/jeremiedb/mxnet_winbin/raw/master/mxnet.zip", repos = NULL)

A bit off subject, but I wonder if there were already some plan to some plans to provide further helpers in R such as bucketing or the multi-threaded mx.image? I could take a look at bucketing to refresh LSTM example, I also noticed the for loop approach for building dictionnary is much less efficient in R than the vectorized approach. Thanks!

[thirdwing]

@jeremiedb Thank you.

As plan, we will try to set up a nightly build for Windows R users in the summer. If you are willing to help update the prebuilt Windows pkg, please let me know.

I know there are lots of work need to be done on R APIs. But as a PhD student, I really don't have much time now.

If you have any suggestion or requirement, please just open an issue. I will close as many issues as possible during my internship this summer.

[thirdwing]

I have upload the latest Windows pkg using libs from @jeremiedb

See https://github.com/dmlc/drat/tree/gh-pages/bin/windows/contrib/3.3

[thsavage]

Thanks to everyone involved. I teach machine learning for social scientists, and despite my preference for Python, R remains the easiest environment to use for people coming from the SAS/Stata world.

MXNET is by far the best deep learning tool out there. You are to be congratulated on your endeavor.

On Sun, Feb 5, 2017 at 1:54 PM Qiang Kou (KK) notifications@github.com wrote:

Closed #4861 https://github.com/dmlc/mxnet/issues/4861.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/dmlc/mxnet/issues/4861#event-949517987, or mute the thread https://github.com/notifications/unsubscribe-auth/AGIm2APKDkoW93rMjZPV8nkmWtLrjVjbks5rZhregaJpZM4L1v9y .

[jeremiedb]

@thirdwing Sure I should be able to provide an update of the prebuilt Windows R package from time to time, at least when there are some significant developments. I actually built from the pre-built Windows binaries (https://github.com/yajiedesign/mxnet/releases), so as long as these are maintained, it should be quite straightforward. I'll contact you directly when I update, if that's ok with you. Thanks!

[thsavage]

Just FYI. This is about a year old, but you might find it of interest.

http://www.rblog.uni-freiburg.de/2017/02/07/deep-learning-in-r/

On Tue, Feb 7, 2017 at 12:46 AM, jeremiedb notifications@github.com wrote:

@thirdwing https://github.com/thirdwing Sure I should be able to provide an update of the prebuilt Windows R package from time to time, at least when there are some significant developments. I actually built from the pre-built Windows binaries (https://github.com/ yajiedesign/mxnet/releases), so as long as these are maintained, it should be quite straightforward. I'll contact you directly when I update, if that's ok with you. Thanks!

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/dmlc/mxnet/issues/4861#issuecomment-277907550, or mute the thread https://github.com/notifications/unsubscribe-auth/AGIm2CM_7GVOqZTS0EIMKCLRWdEkmAG7ks5raAUtgaJpZM4L1v9y .