Repository navigation
Decouple the solvent and vacuum lambda schedules for the AHFE Protocol and remove redundant windows. - #2241
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #2241 +/- ##
==========================================
- Coverage 95.13% 90.37% -4.76%
==========================================
Files 224 224
Lines 22009 22086 +77
==========================================
- Hits 20938 19961 -977
- Misses 1071 2125 +1054
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
|
pre-commit.ci autofix |
| 0.0, 0.25, 0.5, 0.75, 1.0, | ||
| ], | ||
| lambda_vdw=[ | ||
| 0.0, 0.0, 0.0, 0.0, 1.0, |
There was a problem hiding this comment.
We could also do all 0.0, but I thought adding at least a 1.0 would make sure the system is exactly the same (i.e. the parameters in the custom nonbonded are the same) as the end state in solvent.
| # When annihilating sterics, the intramolecular LJ interactions | ||
| # are also turned off, so the vacuum leg needs intermediate vdw windows | ||
| vacuum_lambda_vdw = self.settings.vacuum_lambda_settings.lambda_vdw | ||
| if self.settings.alchemical_settings.annihilate_sterics and not any( |
There was a problem hiding this comment.
Leaving this outside of validate_lambda_schedule since it's very specific to just the vacuum vdw.
|
pre-commit.ci autofix |
| license-files = [ "LICENSE" ] | ||
| authors = [ { name = "The OpenFE developers", email = "openfreeenergy@omsf.io" } ] | ||
| requires-python = ">=3.11" | ||
| requires-python = ">=3.11,<3.15" |
There was a problem hiding this comment.
@atravitz FYI - pre-commit wanted to add py3.15 to the classifiers, so I'm forcing it not to do that.
There was a problem hiding this comment.
FYI - I'm generating a new file, just going to a take a little bit of time.
| @pytest.mark.parametrize( | ||
| "val", | ||
| [ | ||
| {"elec": [0.0, 1.0], "vdw": [1.0, 1.0], "restraints": [0.0, 0.0]}, | ||
| ], | ||
| ) |
There was a problem hiding this comment.
Did we lose a confirguration here for the parametrize flag or should this be removed for:
- test_validate_lambda_schedule_naked_charge
- test_validate_lambda_schedule_nreplicas
- test_validate_lambda_schedule_nwindows
- test_validate_lambda_schedule_nonzero_restraints
There was a problem hiding this comment.
sorry not fully sure what you mean here - is the question about why there's only one entry in the parameterize?
jthorton
left a comment
jthorton
left a comment
There was a problem hiding this comment.
Yeeey more settings, minor test things to maybe change but I'll approve early!
hannahbaumann
left a comment
hannahbaumann
left a comment
There was a problem hiding this comment.
Thanks @IAlibay , lgtm!
|
Going to merge as completed, benchmarks might come in the next 24h or so, but there's issues with HPC3. It shouldn't hold off the release though. |
…l and remove redundant windows. (#2241) * Remove redundant gas phase lambda windows from the AHFE protocol * Decouple gas and solve phase lambda windows in the AHFE protocol settings.
* add --name option to setup-task-campaign * Apply suggestion from @hannahbaumann Co-authored-by: Hannah Baumann <43765638+hannahbaumann@users.noreply.github.com> * Allow reading of legacy restraint settings for SepTop (#2237) * Add fix for issue-2236 * Decouple the solvent and vacuum lambda schedules for the AHFE Protocol and remove redundant windows. (#2241) * Remove redundant gas phase lambda windows from the AHFE protocol * Decouple gas and solve phase lambda windows in the AHFE protocol settings. * main name arg required --------- Co-authored-by: Irfan Alibay <IAlibay@users.noreply.github.com> Co-authored-by: Hannah Baumann <43765638+hannahbaumann@users.noreply.github.com>
3 participants
Description
LLM / AI generated code disclosure
LLMs or other AI-powered tools (beyond simple IDE use cases) were used in this contribution: yes / no
If yes, please provide details here:
Checklist
newsentry, or the changes are not user-facing.pre-commit.ci autofix.Manual Tests: these are slow so don't need to be run every commit, only before merging and when relevant changes are made (generally at reviewer-discretion).
Developers certificate of origin