modelSelection.py 文件源码

def __init__(self, data, mleDiffCutoff=1.0):
        print [min(data), max(data)]

        distributions = [st.laplace, st.norm, st.expon, st.dweibull, st.invweibull, st.lognorm, st.uniform]
        mles = []

        for distribution in distributions:
            pars = distribution.fit(data)
            mle = distribution.nnlf(pars, data)
            mles.append(mle)

        results = [(distribution.name, mle) for distribution, mle in zip(distributions, mles)]

        for dist in sorted(zip(distributions, mles), key=lambda d: d[1]):
            print dist
        best_fit = sorted(zip(distributions, mles), key=lambda d: d[1])[0]
        print 'Best fit reached using {}, MLE value: {}'.format(best_fit[0].name, best_fit[1])          

        self.modelSets = []

        self.modelOptions = [mod[0].name for mod in sorted(zip(distributions, mles), key=lambda d: d[1])]
        ## list of scipy distribution ids sorted by their MLEs given the data
        ## [0] is best, [1], next best and so on


        for model in sorted(zip(distributions, mles), key=lambda d: d[1]):
            if(model[0].name in getAvailableDistributionsByScipyIds()):
                try:
                    modelDist = getDistributionByScipyId(model[0].name, data)
                    self.modelSets.append([modelDist, model[1]])
                    ## append the distribution object and the MLE value for this
                    ## particular distribution & the data

                    ## ah frig, I think in the bimodal case, it will be
                    ## something like 
                except RuntimeError:
                    pass    
            else:
                ## nothing that can be done here, if we dont have a object of
                ## the distribution needed available, we cant do much about it
                pass