此错误表明在`scipy.linalg`中无法导入`pinv2`。`pinv2`在较新版本的`scipy`中已被弃用，替代方案是使用`scipy.linalg.pinv`。以下是解决该问题的方法： 将代码中`from scipy.linalg import pinv2`替换为`from scipy.linalg import pinv`，并将代码里所有`pinv2`替换成`pinv`。 示例代码如下： ```python import numpy as np from sklearn.model_selection import train_test_split from sklearn.metrics import mean_squared_error from scipy.linalg import pinv # 定义核极限学习机类 class KELM: def __init__(self, C=100, kernel_type='RBF', gamma=0.1): self.C = C self.kernel_type = kernel_type self.gamma = gamma def kernel_matrix(self, X1, X2): if self.kernel_type == 'RBF': n1 = X1.shape[0] n2 = X2.shape[0] K = np.zeros((n1, n2)) for i in range(n1): for j in range(n2): diff = X1[i, :] - X2[j, :] K[i, j] = np.exp(-self.gamma * np.dot(diff, diff)) return K else: raise ValueError("Unsupported kernel type") def fit(self, X, y): self.X_train = X K = self.kernel_matrix(X, X) self.beta = pinv(K + np.eye(K.shape[0]) / self.C).dot(y) def predict(self, X): K_test = self.kernel_matrix(X, self.X_train) return K_test.dot(self.beta) # 生成示例数据 np.random.seed(42) n_samples = 1000 n_features = 5 X = np.random.randn(n_samples, n_features) y = np.random.randn(n_samples) # 划分训练集和测试集 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) # 创建KELM模型并训练 kelm = KELM(C=100, kernel_type='RBF', gamma=0.1) kelm.fit(X_train, y_train) # 进行预测 y_pred = kelm.predict(X_test) # 计算均方误差 mse = mean_squared_error(y_test, y_pred) print(f"Mean Squared Error: {mse}") ```