DFTB+ interface upgrade #179
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #179 +/- ##
=======================================
Coverage 91.86% 91.86%
=======================================
Files 47 47
Lines 6748 6748
Branches 758 758
=======================================
Hits 6199 6199
Misses 537 537
Partials 12 12
Flags with carried forward coverage won't be shown. Click here to find out more. |
danielhollas
force-pushed
the
dftb-interface-cleanup
branch
from
d47720a to
59a915b
Compare
danielhollas
force-pushed
the
dftb-interface-cleanup
branch
from
59a915b to
984c710
Compare
There was a problem hiding this comment.
I have not tested the code on my machine. I added general comments, mainly on what to do when people do not work on clusters with the same env as in Prague. Also, might be worth adding a README file for DFTB or at least add somewhere link for documentation and for different parameters.
interfaces/DFTB/dftb_in.hsd
Outdated
| charge = 0.0 | ||
| #ReadInitialCharges = yes | ||
| SlaterKosterFiles = Type2FileNames { | ||
| Prefix = "/home/hollas/3ob-2-1/" | ||
| Prefix = "/home/hollas/3ob-3-1/" |
There was a problem hiding this comment.
Probably add something general with a comment for users, instead of hollas home which will not be present on most clusters.
There was a problem hiding this comment.
I changed this and added a note at the top of r.dftb that this file needs to be modified.
| input=input$ibead | ||
| geom=../geom.dat.$ibead | ||
| if [[ -f ../SetEnvironment.sh ]]; then | ||
| # This is specific to Prague clusters |
There was a problem hiding this comment.
What should people do if they do not use the Prague clusters? Remove SetEnvironment or change it?
There was a problem hiding this comment.
If they are not on Prague clusters this file will simply not exist.
interfaces/DFTB/r.dftb
Outdated
| # This is specific to Prague clusters | ||
| source ../SetEnvironment.sh DFTB | ||
| else | ||
| # We assume dftb+ is in PATH already |
There was a problem hiding this comment.
Some test would be better instead of assumption.
There was a problem hiding this comment.
I opted not to add a check for this, since if it is not in the PATH we will fail to execute anyway, and the default unix error message should be clear enough. I did add a note about this at the top of the file though.
I expanded the comment at the top of the |
| geom=../geom.dat."$ibead" | ||
| natom=$(wc -l < "$geom") | ||
|
|
||
| function prepare_dftb_inputs() { |
There was a problem hiding this comment.
I created two new functions, generate_dftb_inputs and extract_energy_and_gradients, hopefully this makes the high-level flow of the script easier to follow. Happy to hear suggestions how to make this even better.
|
I've tested the refactored interface locally with DFTB+ 24.1 by checking the energy conservation in a short NVE simulation of a single water molecule
So this is now ready for review again. |
No description provided.