error installing package,about google/causalimpact

Comments (48)

steve-the-bayesian commented on May 4, 2024 3

The CRAN maintainers don't have the latest version of BOOM published for linux on x64. The workaround is to download the package source and
MAKEFLAGS="-j 24" R CMD INSTALL Boom_0.8.tar.gz

from causalimpact.

timandrews335 commented on May 4, 2024 1

The CRAN maintainers don't have the latest version of BOOM published for linux on x64. The workaround is to download the package source and
MAKEFLAGS="-j 24" R CMD INSTALL Boom_0.8.tar.gz

Thank you

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Can you try updating the Boom package and see if that resolves the error?

…

On Thu, May 31, 2018 at 7:28 AM, merveakbas ***@***.***> wrote: Hello hello, I upgraded my base R to the latest version yesterday. When I try to re-install causal impact, I keep getting the following error (and previously it was working all fine) Warning: S3 methods ‘ToString.default’, ‘ToString.table’ were declared in NAMESPACE but not found Error : object ‘GammaPrior’ is not exported by 'namespace:Boom' does anyone have a suggested solution for this? Thanks much in advance... — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#22>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1S8DJrXXw6bzyFp0KfDpAMprs5qAks5t3_4AgaJpZM4UVJem> .

from causalimpact.

merveakbas commented on May 4, 2024

I tried

library(Boom) 
library(CausalImpact)```

still getting the same error...

from causalimpact.

steve-the-bayesian commented on May 4, 2024

update.packages("Boom") Restart R. library(CausalImpact)

…

On Thu, May 31, 2018 at 1:11 PM, merveakbas ***@***.***> wrote: I tried library(Boom) library(CausalImpact)``` still getting the same error... — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1SZ8hgavtHWRjjLEzanmgMyfwvPpks5t4E59gaJpZM4UVJem> .

from causalimpact.

merveakbas commented on May 4, 2024

Just tried that - still getting the same error (???)
I have the latest version of Rstudio as well - just double checked that.

from causalimpact.

merveakbas commented on May 4, 2024

I double checked that I have the latest version of Boom as well - it is 0.8

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Does the same thing happen when you library(bsts) bsts is a dependency of CausalImpact, and Boom is a dependency of bsts. If you can load and use bsts then CausalImpact needs updating. If you can't then either bsts or Boom needs updating. The latest bsts is 0.8.0. Are you Windows, Mac, or Linux?

…

On Thu, May 31, 2018 at 1:29 PM, merveakbas ***@***.***> wrote: I double checked that I have the latest version of Boom as well - it is 0.8 — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1WI2-84kspkmpzmefOgm1yh9Ma1Yks5t4FK9gaJpZM4UVJem> .

from causalimpact.

merveakbas commented on May 4, 2024

I think my initial problem is that BoomSpikeSlab is not the latest version

There are binary versions available but the source versions are later:
binary source needs_compilation
BoomSpikeSlab 0.9.0 1.0.0 TRUE
bsts 0.7.1 0.8.0 TRUE

But reinstalling "BoomSpikeSlab" fails with this error:

logit_spike_slab_wrapper.cc:7:10: fatal error: 'cpputil/Ptr.hpp' file not found
#include "cpputil/Ptr.hpp"

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Interesting. The Boom results page on CRAN shows an error for OSX, so I wonder if the OSX version didn't get built. I have asked on r-package-devel.

…

On Thu, May 31, 2018 at 2:09 PM, merveakbas ***@***.***> wrote: I think my problem lies in here There are binary versions available but the source versions are later: binary source needs_compilation BoomSpikeSlab 0.9.0 1.0.0 TRUE bsts 0.7.1 0.8.0 TRUE I answer yes but it somehow doesnt work — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1TG7-D7qvs7qH8Zvx83VkgxdI2u3ks5t4FwigaJpZM4UVJem> .

from causalimpact.

merveakbas commented on May 4, 2024

Thanks much for checking !

from causalimpact.

merveakbas commented on May 4, 2024

For now, I reinstalled R.3.3 and am using those libraries... I look forward to getting news on this though since I know now that the problem is that I can't update my BoomSpikeSlab package

from causalimpact.

steve-the-bayesian commented on May 4, 2024

For testing purposes, how do you know the culprit is BoomSpikeSlab vs Boom or bsts?

…

On Fri, Jun 1, 2018 at 7:58 AM, merveakbas ***@***.***> wrote: For now, I reinstalled R.3.3 and am using those libraries... I look forward to getting news on this though since I know now that the problem is that I can't update my BoomSpikeSlab package — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1cku-QZ3AWawcbF7fABjcpr7VVuHks5t4VaPgaJpZM4UVJem> .

from causalimpact.

merveakbas commented on May 4, 2024

Sorry for late reply: Boom upgrades to latest version successfully, but after that BoomSpikeSlab does not upgrade, it fails. And i believe after that, both btst and Causal Impact failed to upgrade..

from causalimpact.

steve-the-bayesian commented on May 4, 2024

…

On Tue, Jun 5, 2018 at 7:10 AM, merveakbas ***@***.***> wrote: Sorry for late reply: Boom upgrades to latest version successfully, but after that BoomSpikeSlab does not upgrade, it fails. And i believe after that, both btst and Causal Impact failed to upgrade.. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1ftnpwOeZzryrHRj3NP0qV39h02Rks5t5pFlgaJpZM4UVJem> .

from causalimpact.

steve-the-bayesian commented on May 4, 2024

The latest versions of Boom, BoomSpikeSlab, and bsts for OSX are now on CRAN . No explanation yet for the previous failure (maybe a build timeout). @merveakbas, can you see if this solves the issue you were having with CausalImpact? On Wed, Jun 6, 2018 at 9:39 AM Steven Scott <[email protected]> wrote:

…

Working hypothesis is that the Boom build for OSX is timing out on CRAN, so they actually don't have the latest version, and that impedes downstream dependencies BoomSpikeSlab and bsts. I'm in touch with Simon Urbanek (OSX maintainer for cran) about the issue. On Tue, Jun 5, 2018 at 7:10 AM, merveakbas ***@***.***> wrote: > Sorry for late reply: Boom upgrades to latest version successfully, but > after that BoomSpikeSlab does not upgrade, it fails. And i believe after > that, both btst and Causal Impact failed to upgrade.. > > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > <#22 (comment)>, > or mute the thread > <https://github.com/notifications/unsubscribe-auth/AFlR1ftnpwOeZzryrHRj3NP0qV39h02Rks5t5pFlgaJpZM4UVJem> > . >

from causalimpact.

merveakbas commented on May 4, 2024

Ah - thanks much for the update!!! - yeah let me check out and I will report.

from causalimpact.

timandrews335 commented on May 4, 2024

I'm getting the same error with Ptr.hpp, but on Ubuntu 18.04 and a first installation of CausalImpact. I see the issue both on R 3.4 (in the bionic repo) and after adding the official R repo and upgrading to 3.5.

On Windows 10 and R 3.4, CausalImpact installed fine.

from causalimpact.

acorso522md commented on May 4, 2024

Where do I get the relevant package source for linux? Your naming convention here lines up with the R version on CRAN but I can't reuse that one it would seem.

from causalimpact.

xbeta commented on May 4, 2024

I am not sure who owns BoomSpikeSlab. The Github repo is a mirror.

But I face similar problem when installing on my macOS


Error: package or namespace load failed for ‘BoomSpikeSlab’ in dyn.load(file, DLLpath = DLLpath, ...):
 unable to load shared object '/r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so':
  dlopen(/r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so, 6): Symbol not found: __ZN4BOOM9GlobalRng3rngE
  Referenced from: /r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so
  Expected in: flat namespace
 in /r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Sam reached out to me on another thread and I sent him bleeding edge copies of Boom, BoomSpikeSlab, and bsts. The issue is that Boom is a large, compiled package, and the compile sometimes times out on CRAN. When that happens they assume the package is broken and disable it on a platform by platform basis.

…

On Thu, Jan 3, 2019 at 7:21 PM Sam Xiao ***@***.***> wrote: I am not sure who owns BoomSpikeSlab. The Github repo is a mirror. But I face similar problem when installing on my macOS Error: package or namespace load failed for ‘BoomSpikeSlab’ in dyn.load(file, DLLpath = DLLpath, ...): unable to load shared object '/r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so': dlopen(/r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so, 6): Symbol not found: __ZN4BOOM9GlobalRng3rngE Referenced from: /r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so Expected in: flat namespace in /r_libs/BoomSpikeSlab/libs/BoomSpikeSlab.so — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1dVutCRDZN3YEEX109CgVTwRVFtQks5u_qytgaJpZM4UVJem> .

from causalimpact.

xbeta commented on May 4, 2024

@steve-the-bayesian

Thanks for helping, but I still face a few issues similar to this thread about missing header files.

from causalimpact.

xbeta commented on May 4, 2024

@steve-the-bayesian After I run brew uninstall binutils, everything seems to work again :-)

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Yay! I'll disregard your other message about Ptr. That was because you had problems building Boom. Sorry you had all the trouble, but I'm happy you got it working.

…

On Fri, Jan 11, 2019, 9:40 PM Sam Xiao ***@***.*** wrote: @steve-the-bayesian <https://github.com/steve-the-bayesian> After I run brew uninstall binutils, everything seems to work again :-) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1b_5TVZPUYDoLU5oCai5MsWCd2qNks5vCUstgaJpZM4UVJem> .

from causalimpact.

juliamnh commented on May 4, 2024

Hi @steve-the-bayesian,

I also have the issue that CausalImpact won't install Unfortunately the "brew uninstall binutils" doesn't work for me. I get "Error: No such keg:/usr/local/Cellar/binutils"

Do you know how I should go about resolving this issue?

Thanks!

Screenshot 1: trying to install CausalImpact
Screenshot 2: trying to install BoomSpikeLab separately

Note that the package Boom is loaded in my library, but then still these packages won't install.

Screenshot 1:

Screenshot 2:

from causalimpact.

steve-the-bayesian commented on May 4, 2024

The compiler errors you're seeing have to do with the BoomSpikeSlab package not being able to find the C++ code distributed with the Boom package. If you have Boom installed through CRAN (e.g. install.packages("Boom"), called from R). Then you should not be getting this error. By any chance did you install the Boom C++ code as a standalone C++ library? If so then dependencies from CRAN probably won't be able to find it. What platform are you running on? What is the output of the command: /usr/lib/bin/Rscript -e "cat(system.file(package='Boom'))" If you check the compiler command this is where the compiler is looking for Boom header files.

…

On Mon, Feb 4, 2019 at 7:19 AM juliamnh ***@***.***> wrote: Hi @steve-the-bayesian <https://github.com/steve-the-bayesian>, I also have the issue that CausalImpact won't install Unfortunately the "brew uninstall binutils" doesn't work for me. I get "Error: No such keg:/usr/local/Cellar/binutils" Do you know how I should go about resolving this issue? Thanks! Screenshot 1: trying to install CausalImpact Screenshot 2: trying to install BoomSpikeLab separately Note that the package Boom is loaded in my library, but then still these packages won't install. Screenshot 1: [image: image 1] <https://user-images.githubusercontent.com/47326774/52217133-1bea7100-2898-11e9-9d84-db463d2498c5.png> Screenshot 2: [image: image 2] <https://user-images.githubusercontent.com/47326774/52217250-5c49ef00-2898-11e9-87a0-07955d889025.png> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFlR1fZFoB6SF5E9Oqo6F1oMihgyJIHTks5vKE-VgaJpZM4UVJem> .

from causalimpact.

juliamnh commented on May 4, 2024

Thank you Steve! I don't think I installed it as a standalone. Yesterday, restarting the Rsession didn't help, for some reason, a couple more attempts today did the trick. No idea why, but it seems to work now! I'm gonna remember your suggestions if I run into this again (or my coworkers..). Cheers!

from causalimpact.

keniatw commented on May 4, 2024

Coming very late in the discussion, but I am facing the same errors. I successfully installed Boom, but bsts seems to fail every time, so CausalImpact doesn't install. Get the same messages as juliamnh above. Any suggestions so far? Running R and RStudion on Linux Ubuntu 18.04... Ugh!

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Boom and bsts are in the middle of a major update. It would have been better to update them simultaneously, but that's not how things work over at Cran. Give it a few days.

…

On Tue, May 21, 2019, 7:44 AM keniatw ***@***.***> wrote: Coming very late in the discussion, but I am facing the same errors. I successfully installed Boom, but bsts seems to fail every time, so CausalImpact doesn't install. Get the same messages as juliamnh above. Any suggestions so far? Running R and RStudion on Linux Ubuntu 18.04... Ugh! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22?email_source=notifications&email_token=ABMVDVORWNVTKKJOXMKR67DPWQDDXA5CNFSM4FCUS6TKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODV4ERAQ#issuecomment-494422146>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABMVDVLXELJOAJSF2IW66O3PWQDDXANCNFSM4FCUS6TA> .

from causalimpact.

keniatw commented on May 4, 2024

Thank you, steve-the-bayesian!
I have a workshop soon and folks are supposed to install CausalImpact for it (I am a participant, not the instructor), so I will keep my fingers crossed and check from time to time. :)

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Boom and BoomSpikeSlab have made it through the process. I'm happy to provide the .tgz for the bsts update if you need to install manually. Hopefully it won't come to that.

…

On Tue, May 21, 2019 at 8:15 AM keniatw ***@***.***> wrote: Thank you, steve-the-bayesian! I have a workshop soon and folks are supposed to install CausalImpact for it (I am a participant, not the instructor), so I will keep my fingers crossed and check from time to time. :) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22?email_source=notifications&email_token=ABMVDVP6S7FAK4SSDVP6Z5TPWQGXXA5CNFSM4FCUS6TKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODV4HV7I#issuecomment-494435069>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABMVDVOKYKK3PTFRCWHUETDPWQGXXANCNFSM4FCUS6TA> .

from causalimpact.

keniatw commented on May 4, 2024

Yay! I'd love to have it. I tried with the 0.8.0 tar.gz but didn't work either. Do you have an updated version I could try?

from causalimpact.

steve-the-bayesian commented on May 4, 2024

I'll need your actual email address to send the files. Mine is available from the boom and bsts packages.

…

On Tue, May 21, 2019 at 9:20 AM keniatw ***@***.***> wrote: Yay! I'd love to have it. I tried with the 0.8.0 tar.gz but didn't work either. Do you have an updated version I could try? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22?email_source=notifications&email_token=ABMVDVKRRGDKVXTIRVHWA6TPWQOODA5CNFSM4FCUS6TKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODV4N5CA#issuecomment-494460552>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABMVDVMIPRHET26YLIGB54TPWQOODANCNFSM4FCUS6TA> .

from causalimpact.

keniatw commented on May 4, 2024

Hi Steve! Kenia here. I'd appreciate any help with the bsts (and CausalImpact for that matter). Not 100% sure what the underlying problem may be. I get a notification saying "/usr/lib/R/etc/Makeconf:174: recipe for target 'aggregate_time_series.o' failed" when trying to install the bsts. Maybe a problem with the gcc compiler instead? I will try with the tar.gz you'll send to me and go from there. I will keep you posted! Thanks very much. It's very kind of you. Cheers, Kenia [email protected]

…

-- [image: Inspire! Sustainable Development Goals] <https://www.un.org/sustainabledevelopment/blog/2019/05/nature-decline-unprecedented-report/?fbclid=IwAR2NWYrgRci3fKvxQWx18YvqMZqCqgWwVtOV0gW4Ug2DCMKOUF91fYq0-C4> KENIA WIEDEMANN Physicist & Environmental Scientist (857) 919-0284 On Tue, May 21, 2019 at 12:32 PM steve-the-bayesian < [email protected]> wrote:

I'll need your actual email address to send the files. Mine is available from the boom and bsts packages. On Tue, May 21, 2019 at 9:20 AM keniatw ***@***.***> wrote: > Yay! I'd love to have it. I tried with the 0.8.0 tar.gz but didn't work > either. Do you have an updated version I could try? > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub > < #22?email_source=notifications&email_token=ABMVDVKRRGDKVXTIRVHWA6TPWQOODA5CNFSM4FCUS6TKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODV4N5CA#issuecomment-494460552 >, > or mute the thread > < https://github.com/notifications/unsubscribe-auth/ABMVDVMIPRHET26YLIGB54TPWQOODANCNFSM4FCUS6TA > > . > — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#22?email_source=notifications&email_token=AB3JZL42ZDSAT6L7OAQZCSTPWQP3RA5CNFSM4FCUS6TKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODV4O76A#issuecomment-494465016>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AB3JZLZDDF5JZ5UMDHZYEW3PWQP3RANCNFSM4FCUS6TA> .

from causalimpact.

steve-the-bayesian commented on May 4, 2024

FYI bsts version 0.9 was just accepted by CRAN, so it should appear for all platforms within the next 24-48 hours. Dynamic regression is fixed in this version, several new state models have been added, and plotting has been improved. The update to BOOM was needed so I could support multivariate time series, which has been by far the most common feature request. I've got that more-or-less working in the dev version, which I hope to release to the world in a month or so (after a bit more debugging and testing). Sorry for the interruption. On Tue, May 21, 2019 at 9:32 AM Steven Scott <[email protected]> wrote:

…

from causalimpact.

abelborges commented on May 4, 2024

Hi, guys. With the current state of affairs, I was able to install Boom from CRAN but I'm still having trouble with bsts. The error message is similar to @keniatw's:

make: *** [/usr/lib64/R/etc/Makeconf:170: create_shared_state_model.o] Error 1
ERROR: compilation failed for package ‘bsts’

I'm running Fedora 30. Any ideas?

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Abel, There is a Boom/bsts update cycle going on right now. Sorry there is a new one so soon, but there was a pretty serious bug in v 0.9 that needed immediate fixing. Email me and I'll reply with the right set of packages. Steve

…

On Thu, Jun 6, 2019 at 1:33 PM Abel Borges ***@***.***> wrote: Hi, guys. With the current state of affairs, I was able to install Boom from CRAN but I'm still having trouble with bsts. The error message is similar to @keniatw <https://github.com/keniatw>'s: make: *** [/usr/lib64/R/etc/Makeconf:170: create_shared_state_model.o] Error 1 ERROR: compilation failed for package ‘bsts’ I'm running Fedora 30. Any ideas? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22?email_source=notifications&email_token=ABMVDVIYCO44VHYNC246VV3PZFYAPA5CNFSM4FCUS6TKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXECNNI#issuecomment-499656373>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABMVDVKPEZO4WM2H7CJTQTDPZFYAPANCNFSM4FCUS6TA> .

from causalimpact.

abelborges commented on May 4, 2024

Thank you, Steve!

from causalimpact.

yxk4525 commented on May 4, 2024

Hi @steve-the-bayesian : I'm also facing the same issue

make: *** [/usr/lib64/R/etc/Makeconf:170: create_shared_state_model.o] Error 1
ERROR: compilation failed for package ‘bsts’

Can you e-mail me the right set of packages?

Thanks,
Yash
[email protected]

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Hi Yash, You (or anyone else who needs them) can get them here: https://drive.google.com/drive/folders/1cb_oYvYzxJvnnXgSeQhrnkpwV-XVF-nC?usp=sharing Sorry for the inconvenience. It should work itself out in a day or two. Steve

…

On Thu, Jun 6, 2019 at 4:42 PM Yash Karwa ***@***.***> wrote: Hi @steve-the-bayesian <https://github.com/steve-the-bayesian> : I'm also facing the same issue make: *** [/usr/lib64/R/etc/Makeconf:170: create_shared_state_model.o] Error 1 ERROR: compilation failed for package ‘bsts’ Can you e-mail me the right set of packages? Thanks, Yash ***@***.*** — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22?email_source=notifications&email_token=ABMVDVOKBIVWUFQBG5OADZTPZGOH3A5CNFSM4FCUS6TKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXEO6WQ#issuecomment-499707738>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABMVDVK5MQ6N6IHZQUJ4OITPZGOH3ANCNFSM4FCUS6TA> .

from causalimpact.

abelborges commented on May 4, 2024

@steve-the-bayesian It worked like a charm. Thank you once again

from causalimpact.

DeFilippis commented on May 4, 2024

Hey guys, I'm on Mac OS, and I'm getting the error:

#include "cpputil/Ptr.hpp"
         ^~~~~~~~~~~~~~~~~
1 error generated.
make: *** [logit_spike_slab_wrapper.o] Error 1
ERROR: compilation failed for package ‘BoomSpikeSlab’
* removing ‘/Library/Frameworks/R.framework/Versions/3.5/Resources/library/BoomSpikeSlab’**```

Did we ever figure out the solution?

from causalimpact.

DeFilippis commented on May 4, 2024

It works now if I install the packages from source from Steve's Google Drive, but surely that isn't how everybody has to resolve the issue? For future people installing from source, here's the code:

install.packages("Boom_0.9.4.tar.gz", repos = NULL, type="source")
install.packages("BoomSpikeSlab_1.2.2.tar.gz", repos = NULL, type="source")
install.packages("bsts_0.9.3.tar.gz", repos = NULL, type="source")

It appears you need to install the packages in that order due to dependencies.

from causalimpact.

DavideMagno commented on May 4, 2024

Hi, I have tried to install Boom 0.9.5 both on a mac locally and on an Ubuntu 18.04 server as a pre-requisite for CasualImpact.

The Mac installation through CRAN was super smooth and took 5 seconds. On Ubuntu, I managed to install version 0.8 but I get errors for any 0.9.x version.

I report below the latest codes before breaking:
../inst/include/Eigen/src/Core/DenseCoeffsBase.h:650:34: warning: ignoring attributes on template argument ‘Eigen::internal::packet_traits<double>::type {aka __vector(2) double}’ [-Wignored-attributes] return internal::first_aligned<int(unpacket_traits<DefaultPacketType>::alignment),Derived>(m); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ERROR: compilation failed for package ‘Boom’

I have the latest g++ (v 7.5) hence nothing to be done on that side?
Thank you in advance

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Davide, The warning you reported would not cause a compilation failure. My guess is that the compilation error is a few lines up. Can you send more output? The reason this can happen is that the Boom build occurs in parallel, so you got this warning as another branch was failing. Eigen generates a ton of warnings, so I'm not surprised by this one. Steve FYI, I'm in the middle of a month-long struggle with CRAN to get bsts updated. Apologies for the instability while that happens. Steve

…

On Fri, Apr 17, 2020 at 11:43 AM Davide Magno ***@***.***> wrote: Hi, I have tried to install Boom 0.9.5 both on a mac locally and on an Ubuntu 18.04 server as a pre-requisite for CasualImpact. The Mac installation through CRAN was super smooth and took 5 seconds. On Ubuntu, I managed to install version 0.8 but I get errors for any 0.9.x version. I report below the latest codes before breaking: ../inst/include/Eigen/src/Core/DenseCoeffsBase.h:650:34: warning: ignoring attributes on template argument ‘Eigen::internal::packet_traits<double>::type {aka __vector(2) double}’ [-Wignored-attributes] return internal::first_aligned<int(unpacket_traits<DefaultPacketType>::alignment),Derived>(m); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ERROR: compilation failed for package ‘Boom’ I have the latest g++ (v 7.5) hence nothing to be done on that side? Thank you in advance — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABMVDVOW77JXEOE3DRBPHKLRNCPLPANCNFSM4FCUS6TA> .

from causalimpact.

DavideMagno commented on May 4, 2024

Hi Steve, after running the installation in the terminal, I found the following errors:

virtual memory exhausted: Cannot allocate memory cc1plus: out of memory allocating 488 bytes after a total of 4653056 bytes /usr/lib/R/etc/Makeconf:176: recipe for target 'LinAlg/VectorView.o' failed make: *** [LinAlg/VectorView.o] Error 1 make: *** Waiting for unfinished jobs....

Then

/usr/lib/R/etc/Makeconf:176: recipe for target 'Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_3.o' failed make: *** [Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_3.o] Error 1

And finally

/usr/lib/R/etc/Makeconf:176: recipe for target 'Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_2.o' failed make: *** [Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_2.o] Error 1

I checked the RAM availability and seems like I have 3.3GB, is that not enough?

                  total        used        free      shared  buff/cache   available
    Mem:           3.9G        320M        3.4G        628K        173M        3.3G
   Swap:            0B          0B          0B

from causalimpact.

steve-the-bayesian commented on May 4, 2024

Hi Davide, I'm not a compiler expert, but my guess is that your compiler has set aside a certain amount of memory for the "stack" (where automatic variables go, like double, or int). The file that appears to be the issue sets up a very large table, and it may be too large for your stack. Quick work arounds might include (1) look for a compiler flag that will allow you to control the stack size, (2) switch to a different compiler (e.g. from gcc to clang). You have ample physical memory on your machine for the build. I ran into the stack size limitation at Google, which is the reason the table-filling code is split into three parts. Yours is the first case outside of Google where I have seen it be a problem. Searching for stack-related flags in man gcc yields the following promising candidates: -fstack-limit-register=reg -fstack-limit-symbol=sym -fno-stack-limit Generate code to ensure that the stack does not grow beyond a certain value, either the value of a register or the address of a symbol. If a larger stack is required, a signal is raised at run time. For most targets, the signal is raised before the stack overruns the boundary, so it is possible to catch the signal without taking special precautions. For instance, if the stack starts at absolute address 0x80000000 and grows downwards, you can use the flags -fstack-limit-symbol=__stack_limit and -Wl,--defsym,__stack_limit=0x7ffe0000 to enforce a stack limit of 128KB. Note that this may only work with the GNU linker. You can locally override stack limit checking by using the "no_stack_limit" function attribute. Steve

…

On Sat, Apr 18, 2020 at 10:43 AM Davide Magno ***@***.***> wrote: Hi Steve, after running the installation in the terminal, I found the following errors: virtual memory exhausted: Cannot allocate memory cc1plus: out of memory allocating 488 bytes after a total of 4653056 bytes /usr/lib/R/etc/Makeconf:176: recipe for target 'LinAlg/VectorView.o' failed make: *** [LinAlg/VectorView.o] Error 1 make: *** Waiting for unfinished jobs.... Then /usr/lib/R/etc/Makeconf:176: recipe for target 'Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_3.o' failed make: *** [Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_3.o] Error 1 And finally /usr/lib/R/etc/Makeconf:176: recipe for target 'Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_2.o' failed make: *** [Models/Glm/PosteriorSamplers/fill_poisson_mixture_approximation_table_2.o] Error 1 I checked the RAM availability and seems like I have 3.3GB, is that not enough? total used free shared buff/cache available Mem: 3.9G 320M 3.4G 628K 173M 3.3G Swap: 0B 0B 0B — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABMVDVK3UYHNXFJIAMJPKHTRNHRCTANCNFSM4FCUS6TA> .

from causalimpact.

alhauser commented on May 4, 2024

Assuming this is resolved with current bsts versions.

from causalimpact.

error installing package about causalimpact HOT 48 CLOSED

Comments (48)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent