p = 1.0 - (nbits * 2) / float(numeff) #.05 * abs(numeff-2)
else:
p = 0
return bestfit + (.45 / float(numtf)) + (.45 * p)
return bestfit
if __name__ == '__main__':
#log.setLevel(logging.DEBUG)
#random.seed(1234*int(os.getenv('SGE_TASK_ID')))
#p = BooleanProb(evaluate)
evalf = partial(evaluate, nbits=2)
#mapfun = getoutputp0p1)
cfg = loadconfig(parsecmd())
p = BooleanProb(2, evalf)
edw = EvoDevoWorkbench(cfg, p)
#p.eval_ = bindparams(edw.arnconfig, p.eval_)
edw.run()
#f = open('genome.save','w')
#f.write(edw.best.phenotype.code.bin)
#zzf.close
#plot_ = bindparams(edw.arnconfig, plotindividual)
#plot_(edw.best.phenotype)
#genresult = test(ewd.best.circuit, range(2,20))
#print 'Generalization: '
#print genresult
#ok = testmp(phenotype, intinps)
# if ok > bestfit:
# bestfit = ok
# bestout = eff
# #print 'best output index is now ',eff
#phenotype.output_idx = bestout
#else:
#print 'output index is ', phenotype.output_idx
#bestfit = testmp(phenotype, intinps)
bestfit =testmp(phenotype, intinps)
return len(intinps) - bestfit
if __name__ == '__main__':
#p = BooleanProb(evaluate)
evalf = partial(evaluate, nbits=6)
cfg = loadconfig(parsecmd())
#mapfun = getoutputp0p1)
p = Multiplexer(6,evalf)
edw = EvoDevoWorkbench(cfg,p)
#p.eval_ = bindparams(edw.arnconfig, p.eval_)
edw.run()
#f = open('genome.save','w')
#f.write(edw.best.phenotype.code.bin)
#zzf.close
#plot_ = bindparams(edw.arnconfig, plotindividual)
#plot_(edw.best.phenotype)
#genresult = test(ewd.best.circuit, range(2,20))
#print 'Generalization: '
#print genresult