def toSVMProblem(codedSampleSet): # Calculate the unique classes classes = [] for sample in codedSampleSet: classifier = getClassifier(sample) if classifier not in classes: classes.append(classifier) classes.sort() # Use libsvm's data container: return svm_problem([classes.index(i) for i in classes], codedSampleSet), codedSampleSet, codedSampleSet, classes
def create_model(self,datasets,opt,opp,part_ids = None): # Should groups and ngroups be idch ? if part_ids is None: part_ids = datasets.pids trn_d_lbl,trn_lbl,trn_dtst = datasets.mkTrain(part_ids=part_ids) ptrn = svm.svm_problem(trn_lbl,trn_dtst) print "create model ..." #opt = svm.svm_parameter(opt) model = svmutil.svm_train(ptrn,opt) # create saving direcotry #self._mkdir(cnt) # create log files #self._save_log(itest,plbl,pval,cnt) model_name = "%s/model/%s.model" % (self._dir,self._name) svmutil.svm_save_model(model_name, model)
def train2svm_prob(train): ltrain = [l for l in lbl_iter(train)] dtrain = [d for d in data_iter(train)] itrain = [i for i in id_iter(train)] return svm.svm_problem(ltrain,dtrain),itrain