我使用了极限学习机（ELM）模型进行预测，并使用K折交叉验证来验证模型的预测效果。但是执行以下代码后，我得到了这个错误信息：

KeyError: "None of [Int64Index([112, 113, 114, 115, 116, 117, 118, 119, 120, 121,\n            ...\n            550, 551, 552, 553, 554, 555, 556, 557, 558, 559],\n           dtype='int64', length=448)] are in the [columns]"

我该如何解决这个问题？问题出在哪里？代码如下：

 dataset = pd.read_excel("un.xls")        X=dataset.iloc[:,:-1]    y=dataset.iloc[:,-1:]            #----------Scaler----------    scaler = MinMaxScaler()    scaler_X = MinMaxScaler()    X=scaler.fit_transform(X)        #---------------------- Divided the datset----------------------        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=2)        # Splits dataset into k consecutive folds (without shuffling by default).        kfolds = KFold(n_splits=5, random_state=16, shuffle=False)       for train_index, test_index in kfolds.split(X_train, y_train):       X_train_folds, X_test_folds = X_train[train_index], X_train[test_index]       y_train_folds, y_test_folds = y_train[train_index], y_train[test_index]              # put all code in the for loop so that for every set of (X_train_folds, y_train_folds), the model is fitted.       # call predict() for corresponding set of X_test_folds       # put all code in the for loop so that for every set of (X_train_folds, y_train_folds), the model is fitted.       # call predict() for corresponding set of X_test_folds        #----------------------------(input size)-------------    input_size = X_train.shape[1]    hidden_size = 23#---------------------------(To fix the RESULT)-------seed =22   # can be any number, and the exact value does not matternp.random.seed(seed)#---------------------------(weights & biases)------------input_weights = np.random.normal(size=[input_size,hidden_size])biases = np.random.normal(size=[hidden_size])#----------------------(Activation Function)----------def relu(x):   return np.maximum(x, 0, x)#--------------------------(Calculations)----------def hidden_nodes(X):    G = np.dot(X, input_weights)    G = G + biases    H = relu(G)    return H#Output weights output_weights = np.dot(pinv2(hidden_nodes(X_train)), y_train)#------------------------(Def prediction)---------def predict(X):    out = hidden_nodes(X)    out = np.dot(out, output_weights)    return out#------------------------------------(Make_PREDICTION)--------------prediction = predict(X_test_folds)    

错误信息如下：

raise KeyError(f”None of [{key}] are in the [{axis_name}]”)

KeyError: “None of [Int64Index([112, 113, 114, 115, 116, 117, 118, 119, 120, 121,\n …\n 550, 551, 552, 553, 554, 555, 556, 557, 558, 559],\n dtype=’int64′, length=448)] are in the [columns]”

回答：

你应该只使用train_test_split()或KFold()来分割数据。不要同时使用两者

正如KFold()的文档所述：

你应该在KFold.split()中仅使用X。所以使用以下代码：

kfolds = KFold(n_splits=5, random_state=16, shuffle=False)   for train_index, test_index in kfolds.split(X):   X_train_folds, X_test_folds = X[train_index], X[test_index]   y_train_folds, y_test_folds = y[train_index], y[test_index]

此外，删除所有X_train和y_train，因为它们不是必需的。

input_size = X.shape[1]def relu(x):   return np.maximum(x, 0)output_weights = np.dot(pinv2(hidden_nodes(X_train_folds)), y_train_folds)

如果代码仍然因KFold()而导致错误，你应该考虑使用train_test_split()，并用train_test_split()的变量替换KFold()的训练和测试变量

对于train_test_split()：

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=2)input_size = X_train.shape[1]def relu(x):   return np.maximum(x, 0)output_weights = np.dot(pinv2(hidden_nodes(X_train)), y_train)prediction = predict(X_test)