communityDetectionEngine.py 文件源码

def prepareClsuterData(self, graph, Timestep, syllable):
        print "Cluster Time ",Timestep, "Syllable:",syllable
        self.Timestep = Timestep
        distances = copy.deepcopy(nx.to_numpy_matrix(graph))

        samples = [x for x in xrange(len(distances))]
        self.ParsedData = ConsensusClster.DataModel.ParseNormal(distances, samples)
        idlist = [ x.sample_id for x in self.ParsedData.samples ]

        if len(dict.fromkeys(idlist)) != len(idlist):
            raise ValueError, 'One or more Sample IDs are not unique!\n\n\nHere is the data'+str(len(dict.fromkeys(idlist)))+'len'+str(len(idlist))

        if not self.pca_only:
            self._postprocess()
            kwds = []
            self.GlobalCDF = dict()
            self.partition1 = dict()
            self.deltaArea = dict()
            self.area = dict()

            self.newKValues = [2,3,4,5,6,7]

            for i in self.newKValues:
                self.partition1[i] = self.run_cluster(i, self.subsamples, self.subsample_fraction, self.norm_var, kwds)
            k = self.createDeltaArea(self.kvalues)

        if Timestep == 62:
            ChangeInDeltaArea = "ConsensusData/DeltaAreaChange"+str(self.graphWidget.Syllable)+str(4)+"Heatmap.tsv"
            AllClusterData = "ConsensusData/AllClusterings"+str(self.graphWidget.Syllable)+str(4)+"Heatmap.tsv"

            with open(ChangeInDeltaArea, 'w') as outfile:
                kValues = 'day' #K-values
                timeSteps = 'hour' #hour
                value = 'value'
                outfile.write("{}\t{}\t{}\n".format(timeSteps,kValues,value))
                for i,j in self.DeltaAreaTimestep.iteritems():
                    KDict = j
                    # print i, KDict
                    for k,value in KDict.iteritems():
                        outfile.write("{}\t{}\t{}\n".format(i,k,value))

            outfile.close()

            with open(AllClusterData, 'w') as outfile:
                pickle.dump(self.FinalPartiction, outfile)

            print AllClusterData, "is the file name for all data", "look at the K value data and choose Wisely:)"

        if k == 1: 
            partition = dict()
            for i in range(len(distances)):
                partition[i] = 0

            self.partition1[1] = partition
        self.FinalPartiction[self.Timestep] = self.partition1

        return self.partition1[k]