def fit(self): pso.dmn = self.dimensions pso.searchRange = float(1)/float(pso.dmn) pso.maxIterations = 100 s,i,g = pso.particleSwarmOptimize(self.divinecost, True, True) self.weights = s return s,i,g
def learn(search): pso.dmn = len(qds[0]) pso.searchRange = float(search) s,i,g = pso.particleSwarmOptimize(MCC, True, True) f = open('weight.txt', 'w') f.write(str(s)) f.close() return s, i, g
def fit(self): pso.dmn = self.dimensions s, i, g = pso.particleSwarmOptimize(self.costfunction, True, True) self.ar = s[:self.p] self.ma = s[self.p:self.p+self.q] #self.varC = s[self.p+self.q] #self.wnC = s[self.p+self.q+1] #self.error = s[self.p+self.q] print s, i, g return s, i, g
def famaLearn(): print "the year is now " + year logname = 'results/newfamaWeights.txt' f = open(logname, 'a') pso.dmn = len(ratios) ans, i, opt = pso.particleSwarmOptimize(famaFrench, True, True) yearString = str(year) report = "Weights: " + str(ans) + ", Spearman: " + str(opt) famaPlot(ans) f.write("\n") f.write(str(year) + " to " + str(int(year) + 4)) f.write("\n") f.write(report) f.close() return ans
def learn(y): print "the year is now " + year logname = 'results/weights.txt' f = open(logname, 'a') pso.dmn = len(ratios) ans, i, opt = pso.particleSwarmOptimize(calculateFitness, True, True) yearstring = str(year) report = "Weights: " + str(ans) + ", Spearman: " + str(opt) ps = plotFitness(ans) f.write("\n") f.write(yearstring + ", Portfolio Size: " + str(ps) + ", 30 Portfolios") f.write("\n") f.write(report) f.close()
def spearLearn(): global year print "the year is now " + year logname = 'results/avgSpearWeights.txt' f = open(logname, 'a') pso.dmn = len(ratios) ans, i, opt = pso.particleSwarmOptimize(bestAvgSpear, True, True) yearString = str(year) report = "Weights: " + str(ans) + ", Spearman: " + str(opt) f.write("\n") f.write(str(year) + " to " + str(int(year) + 3)) f.write("\n") f.write(report) f.close() return ans