Wading through graphs output by readata.py isnt very time efficient. Would be better for a script to automatically run linear regressions, then regressions on the first set of models against other parameters, build up a bunch of those models, sort the models by r^2/under 0.05 and print to console.
Actually now that I described it, thats probably going to be fairly slow anyways, but it wastes less of my time. Models should probably be their own object, with a base stat to build from, followed by a list of 2 lists with the stat id, followed by its weight.
Wading through graphs output by readata.py isnt very time efficient. Would be better for a script to automatically run linear regressions, then regressions on the first set of models against other parameters, build up a bunch of those models, sort the models by r^2/under 0.05 and print to console.
Actually now that I described it, thats probably going to be fairly slow anyways, but it wastes less of my time. Models should probably be their own object, with a base stat to build from, followed by a list of 2 lists with the stat id, followed by its weight.
Ahh, hmm