def __auto_judge(self, feature):
fv = FeatureExtract.vector_feature(feature)
if self.__clf is not None:
target = self.__clf.predict(fv)[0]
confidence = 100 * max(self.__clf.predict_proba(fv)[0])
else:
target = -1
confidence = 0
return target, confidence
def __auto_judge(self, feature):
fv = FeatureExtract.vector_feature(feature)
if self.__clf is not None:
target = self.__clf.predict(fv)[0]
confidence = 100 * max(self.__clf.predict_proba(fv)[0])
else:
target = -1
confidence = 0
return target, confidence
def __relearn_clf(self,feature,decision):
self.__F.append(FeatureExtract.vector_feature(feature))
self.__L.append(decision)
self.__clf = tree.DecisionTreeClassifier(**self.__dtree_param)
self.__clf.fit(self.__F, self.__L)
def __relearn_clf(self, feature, decision):
self.__F.append(FeatureExtract.vector_feature(feature))
self.__L.append(decision)
self.__clf = tree.DecisionTreeClassifier(**self.__dtree_param)
self.__clf.fit(self.__F, self.__L)
