Add training #125
Add training #125
Conversation
| @@ -192,7 +194,18 @@ | |||
| " reactions = reactionDict[familyName]\n", | |||
| " print 'reactions.py previously has {} rxns. Now adding {} new rxn(s).'.format(len(trainingDatabase.entries.values()), len(reactions))\n", | |||
| " print '================='\n", | |||
| " kineticFamily.saveTrainingReactions(reactions, shortDesc='Training reaction from kinetics library: {0}'.format(libraryName))" | |||
| " for reaction in reactions:\n", | |||
There was a problem hiding this comment.
Reminder to self: I wrote this for loop workaround to solve the issue that family.saveTrainingReactions doesn't seem to allow different meta-data (shortDesc, reference, referenceType, longDesc) per reactions. The code now labels every single reaction with the same index as well as creates an extra line break after every entry.
Some combination of this causes the training reactions to be un-usable without manual edits
Both of these are taken account into the family. saveTrainingReactions. My initial thoughts is to change this function to allow different meta-data rather than trying to debug the work-around.
…brary Specifically changed the way it reads comments, so that it can copy shortDesc and longDesc from the library entry. This is needed in the case where we make the library as an intermediate to making training reactions.
…ating training reactions
|
The convertKineticsLibraryToTrainingReactions.ipynb notebook will be moved to RMG-Py. The changes here have been integrated into ReactionMechanismGenerator/RMG-Py#1373. The createKineticsLibrary.py script has been moved to the Work-In-Progress_scripts repo. |
More scripts to make speed up adding training reactions