多变量线性回归TensorFlow实现--波士顿房价案例

波士顿房价案例

#多元线性回归--波士顿房价问题
#第一步：导入库函数
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from sklearn.utils import shuffle
import pickle as p

%matplotlib inline
#第二步：读入数据
df = pd.read_csv('D:/data/boston.csv',header=0)
#第三步 数据处理
df=np.array(df)
for i in range(12):
    df[:,i]=(df[:,i]-df[:,i].min())/(df[:,i].max()-df[:,i].min())
x_data=df[:,:12]#特征列
y_data=df[:,12]#标签列
#第四步 定义占位符
x=tf.placeholder(tf.float32,[None,12],name='X')
y=tf.placeholder(tf.float32,[None,1],name='Y')
#第五步  定义模型函数 定义命名空间
with tf.name_scope('Model'):
    w=tf.Variable(tf.random_normal([12,1],stddev=0.01),name='W')
    b=tf.Variable(1.0,name='b')
    def model(x,w,b):
        return tf.matmul(x,w)+b
    #定义前向计算
    pred=model(x,w,b)
    
#第六步  设置训练参数
train_epochs = 50
learning_rate = 0.01

#第七步 定义损失函数
with tf.name_scope('LossFunction'):
    loss_function=tf.reduce_mean(tf.pow(y-pred,2))
#第八步： 定义优化器
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss_function)
#第九步： 建立会话
sess=tf.Session()
init = tf.global_variables_initializer()
sess.run(init)

#准备写入tensorboard
logdir='D:\data'
sum_loss_op = tf.summary.scalar('loss',loss_function)#记录损失值，写入tensorboard scalars栏
merged = tf.summary.merge_all()#日志文件合并，方便一次写入
writer = tf.summary.FileWriter(logdir,sess.graph)

#第十步：训练模型
for epoch in range(train_epochs):
    loss_sum=0
    for xs,ys in zip(x_data,y_data):
        xs=xs.reshape(1,12)
        ys=ys.reshape(1,1)
        
        _,summary_str,loss = sess.run([optimizer,sum_loss_op,loss_function],feed_dict={x:xs,y:ys})
        writer.add_summary(summary_str,epoch)
        loss_sum+=loss
    #打乱顺序
    x_data,y_data = shuffle(x_data,y_data)
    b0temp=b.eval(session=sess)
    w0temp=w.eval(session=sess)
    loss_average = loss_sum/len(y_data)
    
    print('epoch=',epoch+1,'loss=',loss_average,'b=',b0temp,'w=',w0temp)
    
n=348
x_test=x_data[n]
x_test=x_test.reshape(1,12)
predict=sess.run(pred,feed_dict={x:x_test})
print('预测值：%f'% predict)
target=y_data[n]
print('标签值：%f'% target)