search

epiforecasts / inc2prev

Estimate epidemiological quantities from repeated cross-sectional prevalence measurements
http://epiforecasts.io/inc2prev/
BSD 3-Clause "New" or "Revised" License
16 stars 7 forks source link

issue when running inc2prev #35

Closed albertosanchezmarroquin closed 2 years ago

albertosanchezmarroquin commented 2 years ago

Hello,

We are trying to run inc3prev within the epi ensamble team. We managed to make it work up to the point when we run the estimate.R -d 1 script, where we get the following error:

ℹ Use spec() to retrieve the full column specification for this data. ℹ Specify the column types or set show_col_types = FALSE to quiet this message. Error: CallrFuture (future_lapply-1) failed. The reason reported was ‘callr subprocess failed: there is no package called ‘inc2prev’’. Post-mortem diagnostic: The parallel worker (PID 5917) started at 2022-04-13T14:14:12+0000 finished with exit code 0. Detected a non-exportable reference (‘externalptr’) in one of the globals (‘future.call.arguments’ of class ‘DotDotDotList’) used in the future expression. The total size of the 9 globals exported is 2.41 MiB. The three largest globals are ‘future.call.arguments’ (2.38 MiB of class ‘list’), ‘...future.FUN’ (18.01 KiB of class ‘function’) and ‘convert_to_weekly’ (10.34 KiB of class ‘function’) Execution halted

Is there any suggestion on what we could try to fix it?

Thank you very much for your help.

Alberto

seabbs commented 2 years ago

Have you been able to use simple-example.R to produce one fit of the model?

Your error message is quite cryptic but it sounds like the parallel process has not found the inc2prev code which is odd as it is loaded in the script.

Do you have devtools installed (responsible for loading the package here)? Otherwise, it might be a package version of environment issue. Where are you running this/with what?

Sam

albertosanchezmarroquin commented 2 years ago

Hi,

Thanks for your response.

In order to run it, we first pull the inc2prev package from your repository, then we mount the code in it and run the scripts in the container. The first two update scripts work fine, but the estimate one fails as above.

We did manage to run the basic example script, and it ran fine, producing the correct output. Yes, we also have devtools installed in the image as we pull it.

Thanks a lot for your help.

Alberto

sbfnk commented 2 years ago

Can you try pulling the latest version and install the R package included using

remotes::install_github("epiforecasts/inc2prev")

and then try running the script again?

sbfnk commented 2 years ago

I assume this has been resolved.