如果不从sklearn中调用呢

如果不想使用scikit-learn库，我们可以使用纯Python和NumPy库来实现葡萄酒分类。以下是一个简单的实现示例：

import numpy as np

# 加载数据集
def load_data():
    data = np.loadtxt('wine.data', delimiter=',')
    X = data[:, 1:] # 特征
    y = data[:, 0] # 标签
    return X, y

# 划分训练集和测试集
def train_test_split(X, y, test_size=0.3, shuffle=True):
    if shuffle:
        # 打乱数据
        perm = np.random.permutation(len(X))
        X = X[perm]
        y = y[perm]

    # 划分训练集和测试集
    split_idx = int(len(X) * (1 - test_size))
    X_train, X_test = X[:split_idx], X[split_idx:]
    y_train, y_test = y[:split_idx], y[split_idx:]

    return X_train, X_test, y_train, y_test

# 计算欧式距离
def euclidean_distance(x1, x2):
    return np.sqrt(np.sum((x1 - x2)**2))

# KNN算法
class KNN:
    def __init__(self, k=3):
        self.k = k

    def fit(self, X, y):
        self.X_train = X
        self.y_train = y

    def predict(self, X):
        y_pred = [self._predict(x) for x in X]
        return np.array(y_pred)

    def _predict(self, x):
        # 计算距离
        distances = [euclidean_distance(x, x_train) for x_train in self.X_train]

        # 取前k个最近邻
        k_idx = np.argsort(distances)[:self.k]
        k_nearest_labels = [self.y_train[i] for i in k_idx]

        # 投票
        most_common = np.argmax(np.bincount(k_nearest_labels))

        return most_common


if __name__ == '__main__':
    X, y = load_data()
    X_train, X_test, y_train, y_test = train_test_split(X, y)

    knn = KNN(k=3)
    knn.fit(X_train, y_train)

    y_pred = knn.predict(X_test)

    accuracy = np.sum(y_pred == y_test) / len(y_test)
    print("准确率为: ", accuracy)

这里我们使用KNN算法作为分类器，通过计算测试样本与训练样本之间的距离，取前k个最近邻，然后采用投票的方式来预测测试样本的类别。最后计算准确率来评估模型的性能。