search

iiasa / RE4AFAGRI_platform

Official repository of the RE4AFAGRI modelling platform, H2020 LEAP-RE
https://www.re4afagri.africa
7 stars 1 forks source link

No outputs written - Zambia replication #16

Open Elgenied opened 1 year ago

Elgenied commented 1 year ago

Dear,

Thank you for your continuous support. I appreciate your assistance in helping me extend the MLED-OnSSET model to other countries that I am currently analyzing.

I have started by replicating the Zambia case study. However, I encountered some challenges during the installation of the R packages, (coming from Python background ) as there were several geospatial packages that have been deprecated since October 2023.

In my particular case, I ran the M-LED_hourly.R file for Zambia, and the log.txt file showed all the steps up until "## starting writing outputs," but the outputs were not actually written. I am working on Windows 10 with R version 4.3, using RStudio.

Elgenied commented 1 year ago

image These are the only results hown in mled/results file . the zambia file is empty.

giacfalk commented 1 year ago

Dear Elgenied, thanks for your interest in RE4AFAGRI and M-LED. Could you please indicate if particular warnings were reported in the RStudio console during the outputs writing phase, and if anything particular is reported when prompting library(sf) into RStudio?

Elgenied commented 1 year ago

the first analyysis i guess because of memory issues , I was allowing Paralell operation to be True , this occured in the last step of writing the outpus after running for several hours , and got this error :

Error: cannot allocate vector of size 874 Kb In addition: There were 50 Warnings

Then I changed the boolen variable to F and it failed at step 10 , this was the error message :

At core/core/vector.pmt:1030 : Assertion failed: v->stor_begin != NULL. This is an unexpected igraph error; please report this as a bug, along with the steps to reproduce it. Please restart your R session to avoid crashes or other surprising behavior. In addition: There were 50 or more warnings (use warnings() to see the first 50) Called from: add_edges(res, edges = t(as.matrix(el[, 1:2])), attr = weight)

I think it might be realted to 'igraph' tool. is there specific version of it youu'd recommend?

giacfalk commented 1 year ago

The first error is indeed a memory error given by the excessive use of resources (could you post the specs of your computer? How much RAM do you have?). Then, I am not familiar with this second error, I recommend restarting a clear RStudio session, and typing a remove.packages("igraph"), followed by an `"install.packages("igraph")``. After that, I would recommend re-running the model. Please let me know

Elgenied commented 1 year ago

Thank you for your answer . regarding my specs it's 16 G RAM , Intel Core i7-10 G , the Laptop isn't at it's optimal condtions so i would expect lower perfromance. I'll re-install the 'igrpah' package and re-run the workflow !

Thanks again for your answer

giacfalk commented 1 year ago

Thanks. It's certainly not an hardware issue then, as the model run on computers with much lower specs.

giacfalk commented 1 year ago

@Elgenied any news?

szhorvat commented 9 months ago

At core/core/vector.pmt:1030 : Assertion failed: v->stor_begin != NULL. This is an unexpected igraph error; please report this as a bug, along with the steps to reproduce it.

When you see a message like this, it is always a bug in igraph. This particular bug may already be fixed in igraph, which received a major update recently. Please test the latest version first (2.0.2 will be on CRAN soon).

If the issue is still present, please try to produce a minimal reproducer that does not depend on any other package than igraph, and open an issue in the R/igraph repo at https://github.com/igraph/rigraph/

giacfalk commented 9 months ago

At core/core/vector.pmt:1030 : Assertion failed: v->stor_begin != NULL. This is an unexpected igraph error; please report this as a bug, along with the steps to reproduce it.

When you see a message like this, it is always a bug in igraph. This particular bug may already be fixed in igraph, which received a major update recently. Please test the latest version first (2.0.2 will be on CRAN soon).

If the issue is still present, please try to produce a minimal reproducer that does not depend on any other package than igraph, and open an issue in the R/igraph repo at https://github.com/igraph/rigraph/

Many thanks for cross-posting this, might be useful for M-LED users!