我试图结合使用堆栈的两种机器学习算法来获得更好的结果,但是在某些方面却失败了。 这是我的代码:
类Ensemble(threading.Thread): “与三个分类模型堆叠以提高预测的准确性” def init (自身,X,Y,XT,YT,accLabel = None): threading.Thread。 init (自己) 自我X = X 自我Y = Y self.XT = XT self.YT = YT self.accLabel = accLabel
def Stacking(self,model,n_fold,train,test,y):
folds=StratifiedKFold(n_splits=n_fold,random_state=1)
test_pred=np.empty((test.shape[0],1),float)
train_pred=np.empty((0,1),float)
for train_indices,val_indices in folds.split(train,y):
x_train,x_val=train.iloc[train_indices],train.iloc[val_indices]
y_train,y_val=y.iloc[train_indices],y.iloc[val_indices]
model.fit(X=x_train,y=y_train)
train_pred=np.append(train_pred,model.predict(x_val))
test_pred=np.append(test_pred,model.predict(test))
return test_pred.reshape(-1,1),train_pred
def run(self):
X = np.zeros(self.X.shape)
Y = np.zeros(self.Y.shape)
XT = np.zeros(self.XT.shape)
YT = np.zeros(self.YT.shape)
np.copyto(X, self.X)
np.copyto(Y, self.Y)
np.copyto(XT, self.XT)
np.copyto(YT, self.YT)
model1 = tree.DecisionTreeClassifier(random_state=1)
n_fold=4
test_pred1 ,train_pred1=self.Stacking(model1, n_fold, X, XT, Y)
train_pred1=pd.DataFrame(train_pred1)
test_pred1=pd.DataFrame(test_pred1)
model2 = KNeighborsClassifier()
test_pred2 ,train_pred2=self.Stacking(model2, n_fold, X, XT, Y)
train_pred2=pd.DataFrame(train_pred2)
test_pred2=pd.DataFrame(test_pred2)
df = pd.concat([train_pred1, train_pred2], axis=1)
df_test = pd.concat([test_pred1, test_pred2], axis=1)
model = LogisticRegression(random_state=1)
model.fit(df,Y)
sd = model.score(df_test, YT)
acc = (sum(sd == YT) / len(YT) * 100)
print("Accuracy of Ensemble Learning Model is : %.2f" % acc+' %')
print('=' * 100)
if self.accLabel: self.accLabel.set("Accuracy of Ensembelance Learning: %.2f" % (acc)+' %')
错误发生在Stacking方法的'iloc'中。
我一直在不断得到np.ndarray的错误,没有属性'iloc'。我尝试搜索,但是找不到任何特定的链接,尽管我认为这与属于np.ndarray的iloc有关。 如果有人可以帮我这个忙!
答案 0 :(得分:0)
正如评论所建议的,.iloc
是熊猫数据框方法。
要过滤一个numpy数组,您只需要:array[indices]
在您的情况下:
x_train,x_val=train[train_indices],train[val_indices]
y_train,y_val=y[train_indices],y[val_indices]