Perceptron

# sign function
def sign(v):
  if v > 0:
    return 1
  else:
    return -1
    
#Define main function
def Perceptron(pos_sample, neg_sample, learning_rate = 1, iter = 5):

    train_datas = pos_sample + neg_sample  # 样本集

    weight = [0, 0]  # 权重
    bias = 0  # 偏置量
    learning_rate = learning_rate  # 学习速率

    train_num = iter  # 迭代次数

    for i in range(train_num):
        train = random.choice(train_datas)
        x1, x2, y = train
        predict = sign(weight[0] * x1 + weight[1] * x2 + bias)  # 输出
        print("train data: x: (%d, %d) y: %d  ==> predict: %d" % (x1, x2, y, predict))
        if y * predict <= 0:  # 判断误分类点
            weight[0] = weight[0] + learning_rate * y * x1  # 更新权重
            weight[1] = weight[1] + learning_rate * y * x2
            bias = bias + learning_rate * y  # 更新偏置量
            print("update weight and bias: "),
            print(weight[0], weight[1], bias)

    print("trained weights and bias: "),
    print(weight[0], weight[1], bias)

    return weight, bias

train_data1 = [[1, 3, 1], [2, 5, 1], [3, 8, 1], [2, 6, 1]]  # 正样本/positive sample
train_data2 = [[3, 1, -1], [4, 1, -1], [6, 2, -1], [7, 3, -1]]  # 负样本/negative sample


weight, bias = Perceptron(train_data1, train_data2, learning_rate = 0.4, iter = 5)