线性回归 tensorflow代码实现

import tensorflow as tf


def linerRession():
    """
    自实现线性回归
    :return: None
    """
    # 1）准备好数据集：y = 0.8x + 0.7 100个样本
    # 特征值X, 目标值y_true
    X = tf.random_normal(shape=(100, 1), mean=3, stddev=2)
    # y_true [100, 1]
    # 矩阵运算 X（100， 1）* （1, 1）= y_true(100, 1)
    y_true = tf.matmul(X, [[0.8]]) + 0.7
    # 2）建立线性模型：
    # y = W·X + b，目标：求出权重W和偏置b
    # 3）随机初始化W1和b1
    weights = tf.Variable(initial_value=tf.random_normal(shape=(1, 1)))
    bias = tf.Variable(initial_value=tf.random_normal(shape=(1, 1)))

    y_predict = tf.matmul(X, weights) + bias
    # 4）确定损失函数（预测值与真实值之间的误差）-均方误差
    error = tf.reduce_mean(tf.square(y_predict - y_true))

    # 5）梯度下降优化损失：需要指定学习率（超参数）
    optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01).minimize(error)

    init = tf.global_variables_initializer()
    with tf.Session() as sess:
        # 运行初始化变量Op
        sess.run(init)
        print("随机初始化的权重为%f， 偏置为%f" % (weights.eval(), bias.eval()))
        # 训练模型
        for i in range(100):
            sess.run(optimizer)
            print("第%d步的误差为%f，权重为%f， 偏置为%f" % (i, error.eval(), weights.eval(), bias.eval()))

    return None

if __name__ == '__main__':
    linerRession()