TensorFlow实现简单线性回归

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

# 屏蔽warning以下的日志信息
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

def generate_data():
    x = tf.constant(np.array([i for i in range(0, 100, 5)]).reshape(-1, 1), tf.float32)
    y = tf.add(tf.matmul(x, [[1.3]]) + 1, tf.random_normal([20, 1], stddev=30))
    return x, y

def myregression():
    """
    自实现线性回归
    :return:
    """
    x, y = generate_data()
    #     建立模型  y = x * w + b
    # w 1x1的二维数据
    w = tf.Variable(tf.random_normal([1, 1], mean=0.0, stddev=1.0), name='weight_a')
    b = tf.Variable(0.0, name='bias_b')

    y_predict = tf.matmul(x, a) + b

    # 建立损失函数
    loss = tf.reduce_mean(tf.square(y_predict - y))
    
    # 训练
    train_op = tf.train.GradientDescentOptimizer(0.1).minimize(loss=loss)

    # 初始化全局变量
    init_op = tf.global_variables_initializer()

  
    with tf.Session() as sess:
        sess.run(init_op)
        print('初始的权重:%f偏置值：%f' % (a.eval(), b.eval()))
    
        # 训练优化
        for i in range(1, 100):
            sess.run(train_op)
            print('第%d次优化的权重:%f偏置值：%f' % (i, a.eval(), b.eval()))
        # 显示回归效果
        show_img(x.eval(), y.eval(), y_predict.eval())

def show_img(x, y, y_pre):
    plt.scatter(x, y)
    plt.plot(x, y_pre)
    plt.show()

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

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'


def generate_data():
    x = tf.constant(np.array([i for i in range(0, 100, 5)]).reshape(-1, 1), tf.float32)
    y = tf.add(tf.matmul(x, [[1.3]]) + 1, tf.random_normal([20, 1], stddev=30))
    return x, y


def myregression():
    """
    自实现线性回归
    :return:
    """
    x, y = generate_data()
    # 建立模型  y = x * w + b
    w = tf.Variable(tf.random_normal([1, 1], mean=0.0, stddev=1.0), name='weight_a')
    b = tf.Variable(0.0, name='bias_b')

    y_predict = tf.matmul(x, w) + b

    # 建立损失函数
    loss = tf.reduce_mean(tf.square(y_predict - y))
    # 训练
    train_op = tf.train.GradientDescentOptimizer(0.0001).minimize(loss=loss)

    init_op = tf.global_variables_initializer()

    with tf.Session() as sess:
        sess.run(init_op)
        print('初始的权重:%f偏置值：%f' % (w.eval(), b.eval()))
        # 训练优化
        for i in range(1, 35000):
            sess.run(train_op)
            print('第%d次优化的权重:%f偏置值：%f' % (i, w.eval(), b.eval()))
        show_img(x.eval(), y.eval(), y_predict.eval())


def show_img(x, y, y_pre):
    plt.scatter(x, y)
    plt.plot(x, y_pre)
    plt.show()


if __name__ == '__main__':
    myregression()