Initial SOMD2 prototype #4
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…PU or CPU resources
…cifically the platform option
…l as device flag in map (it's case sensitive)
… metadata (for now just temperature and lambda value) to outputs, and cleaned up some unneeded duplication
Analysis to be added in separate module for greater flexibility. _sire_merge_runsim now returns the raw energy trajectory as a pandas dataframe.
_analysis.py currently just finds gradients, reduced energies and forward and backward MC, doing so for all energy trajectories in a given folder, formatting them in to a list of dataframes for analysis with alchemlyb (not yet added)
Added handling of errors such that the whole process doesn't crash if a single simulation crashes - still very rudimentary, seems to only capture the last exception that occurs Removed some redundant code, move dataframe_to_parquet inside of the controller class
Issues with openmm minimisation was causing certain lambda windows to fail outright, now added ufnctionality in which failed minimisation will be attempted again at lambda=0. Also added improved logging - still very much a WIP
analyse_all now returns a list of dataframes sorted by lambda value from smallest to largest, also removed columns with nan from equilibration
Options are now set by calling create_sim_options after initialising a controller object, passing in a dictionary of key value pairs, if options are not set they will default to pre-selected values, to see these values, as well as a full dictionary of all currently supported options, get_defaults can be called
…ssure is given by the user
Both parquet files and trajectories will be saved to user-specified output directory, if not specified they will be saved to the current working directory
More coherent logging is something that needs to be revisisted in the future, this is just a hacky version
…value. This also fixes the issue with saving trajectories
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR brings the initial SOMD2 prototype into
main. The command-line interface is designed to be near identical to SOMD1 so that it can be used as a drop-in replacement, allowing easy comparison.Outstanding things that are not yet supported:
standard_morphandcharge_scaled_morphas keywords. More complex implementations require direct scripting. We might want to think about a separate configuration format for lambda schedules.This PR supersedes and closes #3.