cvignac
commented
2 months ago 500Gb should be more than enough. The segmentation fault refers to a memory access issue, but it does not necessarily mean that not enough memory is available.
Make sure that rdkit is imported at the first line of main.py, it helped me in the past. Else, check that all dependency versions match the requirements file.
Clement
On Tue, Apr 30, 2024 at 7:07 AM SophieSarceau @.***> wrote:
When I use the default setting (qm9 dataset) for model training. During the data preprocessing stage I encounted a problem that the compute node core dumped caused by insufficient memory. image.png (view on web) https://github.com/cvignac/DiGress/assets/46676385/18b9bc9a-37a6-4f49-bf02-b7a65257b8a8
I used a A800 Sserver with 500GB Memory. I wonder how much memory is required for running DiGress? Thanks for your help!
— Reply to this email directly, view it on GitHub https://github.com/cvignac/DiGress/issues/94, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEJOOTVWCF6NESHSLDRUEILY74YJTAVCNFSM6AAAAABG7S6LYSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI3TANJYGIZTSMI . You are receiving this because you are subscribed to this thread.Message ID: @.***>
When I use the default setting (qm9 dataset) for model training. During the data preprocessing stage I encounted a problem that the compute node core dumped caused by insufficient memory.
I used a A800 Sserver with 500GB Memory. I wonder how much memory is required for running DiGress? Thanks for your help!