Python framework for writing programs to solve complex problems not amenable to brute force. The architecture is a descendant of Douglas Hofstadter and Melanie Mitchell's Copycat architecture.
Currently, when saving an LTM, the last state within current run is saved. It would be better to be able to push only the salient items and edges out instead of the whole graph.
This requires some rejiggering (mostly of LTM manager, but tweaking other LTM related classes is also a good idea)
Currently, when saving an LTM, the last state within current run is saved. It would be better to be able to push only the salient items and edges out instead of the whole graph.
This requires some rejiggering (mostly of LTM manager, but tweaking other LTM related classes is also a good idea)