如何修复标签数量与样本数量不匹配的问题？

Question

我正在尝试运行随机森林分类器，在 test_train_split 之后，当我运行它时会生成一个错误，即样本数与样本不匹配。我有 960 行和 26 列。X=pd.read_csv('csv',delimiter=',',sep=",",quotechar='"',header=None,quoting=2,error_bad_lines=False,na_values="")X=X.drop([0,2,4,5],axis=1)X= X.fillna('')X= X[0:961].astype("int")features= X.columnsY= X[27] #my labelsY= Y[0:961].astype("int") # Converting datatype string to intY=Y[np.isfinite(X[27])] # Remove NAN values from my label Yx_train,y_train,x_test,y_test=train_test_split(X,Y,test_size=0.3)clf = RandomForestClassifier(n_estimators=50,max_depth=3)clf.fit(x_train,y_train)