Method predict

MachineLearningAlgorithm/KNN/KNN.py:119–172

(self, X_p)

    """

    def predict(self, X_p):
        k = 11  # 超参数取11

        predict_true = 0  # the num of right predicted
        max = X_p.shape[0] # the max num of iteration
        y_p = []
        for i in range(max):
            x_p = X_p[i]

            distances = [np.sqrt(np.sum((x_p - x) ** 2)) for x in self.X_train]
            # calculate the distance between point in x_p and point in x
            d = np.sort(distances)
            # sort the distances
            nearest = np.argsort(distances)
            # the index of sorted data
            # print(nearest)

            topk_y = [self.y_train[j] for j in nearest[:k]]
            # select k nearest num

            for i in range(k):
                if topk_y[i] == 'pear':
                    color = 'b'
                elif topk_y[i] == 'poplar':
                    color = 'y'
                else:
                    color = 'g'
                plt.scatter(self.X_train[nearest[i], 0], self.X_train[nearest[i], 1], color=color)
            classCount = {}
            for i in range(0, k):
                voteLabel = topk_y[i]
                weight = gaussian(distances[nearest[i]])
                # print(index, dist[index],weight)
                ## 这里不再是加一，而是权重*1
                classCount[voteLabel] = classCount.get(voteLabel, 0) + weight * 1

            maxCount = 0

            for key, value in classCount.items():
                if value > maxCount:
                    maxCount = value
                    classes = key
            # select the type of max num
            y_p.append(classes)
            if topk_y[i] == 'pear':
                color = 'b'
            elif topk_y[i] == 'poplar':
                color = 'y'
            else:
                color = 'g'
            plt.scatter(x_p[0], x_p[1], color=color, s=100, edgecolors='r')
        # plt.savefig("./Pictures/result.png")  # 保存原始数据分布图

        return y_p


    def visualize(self, ssler, x, y) :