TensorFlow 模拟線性回歸代碼樣例

import numpy as np
    import tensorflow as tf
    import matplotlib.pyplot as plt

    #1 構造待模拟線性回歸的點
    num_points = 
    vector_set = []
    for i in range(num_points):
        x1 = np.random.normal(, )
        y1 = x1 *  +  + np.random.normal(, )
        vector_set.append([x1, y1])
    x_data = [v[] for v in vector_set]
    y_data = [v[] for v in vector_set]
    #2 定義計算架構
    ##2.1 定義拟合函數
    w = tf.Variable(tf.random_uniform([], -, ), name="w")
    b = tf.Variable(tf.zeros([]), name="b")
    y = w * x_data + b

    ##2.2 定義loss函數
    loss = tf.reduce_mean(tf.square(y - y_data) * , name="lose")
    ##2.3 定義優化器梯度下降法
    optimizer = tf.train.GradientDescentOptimizer()
    ##2.4 定義訓練目标
    train = optimizer.minimize(loss, name='train')

    #3 疊代計算進行訓練
    init = tf.global_variables_initializer()
    with tf.Session() as sess:
        sess.run(init)
        print("w=", sess.run(w), ", b=", sess.run(b), ", loss=", sess.run(loss))

        for _ in range():
            sess.run(train)
            print("w=", sess.run(w), ", b=", sess.run(b), ", loss=", sess.run(loss))
        ##這裡增加了一個步驟，将拟合出的函數顯示到坐标圖上
        plt.scatter(x_data, y_data, c="r")
        plt.plot(x_data, sess.run(w) * x_data + sess.run(b))
    plt.show()
           

以上代碼出處為以下課程，本人在完成該課程學習時，在本地環境運作了以上代碼并在注釋中增加了自己的了解。

http://study.163.com/course/courseMain.htm?courseId=1003606092