Closed ZacZZZ closed 4 years ago
@ZacZZZ it is still something to do with old versions of DSC ... Are you using the downloaded database I provide, or is this generated from your own runs? If it is the former case I'll see if I can fix the database manually for susie-paper
results -- I dont want to replace it because then the docker image version will not work.
I updated dsc from 0.3.11 to 0.3.12 and the problem persists. Do you refer to my own data when saying "your own runs"? I am still following the instruction you posted online. For the dsc, you might also want to update dsc install page as conda still uses the 0.3.11 version. When I run the docker file manually some of the versions you specified seems cannot be installed
@ZacZZZ Okay version 0.3.12 is fine -- it is just that for that file the new version no longer works with what was generated under a previous version. Can you zip and post here the file called susie_comparison/susie_comparison.db
? I'll see if I can manually fix it for you and get it back to you. When I do, I'll move this ticket to susie-paper
and pin you there.
@gaow susie_comparison_db.zip
Here you go. [Edit: I just realized the zip command in Linux also incorporates the path. I have updated the zip file]
@ZacZZZ see here https://github.com/stephenslab/susie-paper/issues/2
Dear Stephenslab,
Thank you for your help so far. I have been making progress toward reproducing your results. I ran into a new error when running you numerical comparison on
And I realized that the working directory is set to ~/GIT/github/mvarbvs/dsc at some step.
Any suggestion?