给以上代码每行添加注释

以下是每行代码的注释：

import numpy as np

# 生成单层决策树的函数
def buildStump(dataArr, classLabels, D):
    # 将数据和标签转换为 NumPy 矩阵
    Matrix = np.mat(dataArr)
    label = np.mat(classLabels).T
    # 获取数据集的行数和列数
    m, n = Matrix.shape
    # 定义步数和初始化最佳单层决策树、最小误差和预测结果
    Steps = 20.0
    bestStump = {}
    bestClasEst = np.mat(np.zeros(shape=(m, 1)))
    minError = np.inf
    # 遍历每个特征
    for i in range(n):
        # 计算该特征的最小值和最大值
        rangeMin = Matrix[:, i].min()
        rangeMax = Matrix[:, i].max()
        # 计算步长
        stepSize = (rangeMax - rangeMin) / Steps
        # 遍历每个步长
        for j in range(-1, int(Steps) + 1):
            # 对于每个不等式符号
            for inequal in ['lt', 'gt']:
                # 计算阈值
                threshVal = (rangeMin + float(j) * stepSize)
                # 根据阈值和不等式符号生成预测结果
                predictedVals = stumpClassify(Matrix, i, threshVal, inequal)
                # 计算误差向量
                errArr = np.mat(np.ones(shape=(m, 1)))
                errArr[predictedVals == label] = 0
                # 计算加权误差率
                weightedError = D.T * errArr
                # 如果加权误差率更小，则更新最佳单层决策树、最小误差和预测结果
                if weightedError < minError:
                    minError = weightedError
                    bestClasEst = predictedVals.copy()
                    bestStump['dim'] = i
                    bestStump['thresh'] = threshVal
                    bestStump['ineq'] = inequal
    # 返回最佳单层决策树、最小误差和预测结果
    return bestStump, minError, bestClasEst

# 训练 AdaBoost 模型的函数
def adaBoostTrainDS(dataArr, classLabels, numIt=40):
    # 将数据转换为 NumPy 矩阵
    dataArr = np.mat(dataArr)
    # 定义弱分类器集合、样本数和初始化权重向量、预测结果
    weakClassArr = []
    m = dataArr.shape[0]
    D = np.mat(np.ones(shape=(m, 1)) / m)
    aggClassEst = np.mat(np.zeros(shape=(m, 1)))
    # 遍历每个迭代
    for i in range(numIt):
        # 生成最佳单层决策树
        bestStump, error, classEst = buildStump(dataArr, classLabels, D)
        # 计算该决策树的权重
        alpha = float(0.5 * np.log((1 - error) / max(error, 1e-16)))
        bestStump['alpha'] = alpha
        # 将该决策树及其权重添加到弱分类器集合中
        weakClassArr.append(bestStump)
        # 计算样本权重的指数项
        expon = np.multiply(-1 * alpha * np.mat(classLabels).T, classEst)
        # 更新样本权重向量并归一化
        D = np.multiply(D, np.exp(expon))
        D = D / D.sum()
        # 计算目前为止模型在训练集上的预测误差
        aggClassEst += alpha * classEst
        aggErrors = np.multiply(np.sign(aggClassEst) != np.mat(classLabels).T, np.ones(shape=(m, 1)))
        errorRate = aggErrors.sum() / m
        print("total error: ", errorRate)
        # 如果模型已经完美地分类了所有样本，则停止训练
        if errorRate == 0.0:
            break
    # 返回弱分类器集合和模型的预测结果
    return weakClassArr, aggClassEst