You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When running the function compile_pseudo_pop on my Linux server, it generates an error message 'Aborted (core dumped)'. However, the same code works on my Windows 10 machine. I am not sure what is the cause of the issue but I suspect that this occurs during the parallelization process (perhaps issue on the nthread argument / number of cores attributed?).
I use R 4.3.0, Platform: x86_64-pc-linux-gnu (64-bit), CausalGPS ‘0.4.1’, and Ubuntu 22.04.2 LTS. I have attached a zip file that contains files with more details about the specifications of my server (machine_info.txt), R code and output (Rcode_and_Routput.txt), and data input (ACLED_1_type.Rdata) here:
I have just found that if I decrease the number of DT threads as follows: data.table::setDTthreads(percent=30) before running the function compile_pseudo_pop, it seems to solve the issue. But percentages higher than 45% seems to lead to the issue previously reported. Perhaps one should further investigate how the function compile_pseudo_pop allocates cores for parallel processes on a Linux machine with particular specification such as mine (AMD Epyc 7713 dual core, 64 x 2 physical cores, 128 x 2 logical cores).
When running the function
compile_pseudo_pop
on my Linux server, it generates an error message 'Aborted (core dumped)'. However, the same code works on my Windows 10 machine. I am not sure what is the cause of the issue but I suspect that this occurs during the parallelization process (perhaps issue on the nthread argument / number of cores attributed?).I use R 4.3.0, Platform: x86_64-pc-linux-gnu (64-bit), CausalGPS ‘0.4.1’, and Ubuntu 22.04.2 LTS. I have attached a zip file that contains files with more details about the specifications of my server (machine_info.txt), R code and output (Rcode_and_Routput.txt), and data input (ACLED_1_type.Rdata) here:
CausalGPS_issue.zip
Many thanks for your support!
Best wishes,
Andre
The text was updated successfully, but these errors were encountered: