学习完基本操作后,今天来学习一下如何用tensorflow建立线性回归模型。
一、首先建立一些数据
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
# 随机生成1000个点,围绕在y=0.1x+0.3的直线周围
num_points = 1000
vectors_set = []
for i in range(num_points):
x1 = np.random.normal(0.0, 0.55)
y1 = x1 * 0.1 + 0.3 + np.random.normal(0.0, 0.03)
vectors_set.append([x1, y1])
# 生成一些样本
x_data = [v[0] for v in vectors_set]
y_data = [v[1] for v in vectors_set]
plt.scatter(x_data,y_data,c='r')
plt.show()
运行结果是:
二、建立模型:
# 生成1维的W矩阵,取值是[-1,1]之间的随机数
W = tf.Variable(tf.random_uniform([1], -1.0, 1.0), name='W')
# 生成1维的b矩阵,初始值是0
b = tf.Variable(tf.zeros([1]), name='b')
# 经过计算得出预估值y
y = W * x_data + b
# 以预估值y和实际值y_data之间的均方误差作为损失
loss = tf.reduce_mean(tf.square(y - y_data), name='loss')
# 采用梯度下降法来优化参数
optimizer = tf.train.GradientDescentOptimizer(0.5)
# 训练的过程就是最小化这个误差值
train = optimizer.minimize(loss, name='train')
sess = tf.Session()
init = tf.global_variables_initializer()
sess.run(init)
# 初始化的W和b是多少
print ("W =", sess.run(W), "b =", sess.run(b), "loss =", sess.run(loss))
# 执行20次训练
for step in range(20):
sess.run(train)
# 输出训练好的W和b
print ("W =", sess.run(W), "b =", sess.run(b), "loss =", sess.run(loss))
writer = tf.summary.FileWriter("./tmp", sess.graph)##writer = tf.train.SummaryWriter(("./tmp", sess.graph))
#这是因为在1.0版本中,tf.train.SummaryWriter已经改为tf.summary.FileWriter
运行结果是:
W = [0.04677415] b = [0.] loss = 0.09138401 W = [0.06286795] b = [0.2992881] loss = 0.0013578985 W = [0.0743912] b = [0.29931584] loss = 0.0011342986 W = [0.08227211] b = [0.29933572] loss = 0.0010297119 W = [0.08766197] b = [0.2993493] loss = 0.0009807928 W = [0.09134818] b = [0.29935864] loss = 0.00095791137 W = [0.09386922] b = [0.29936498] loss = 0.00094720896 W = [0.09559341] b = [0.29936934] loss = 0.0009422029 W = [0.0967726] b = [0.29937232] loss = 0.00093986175 W = [0.09757906] b = [0.29937434] loss = 0.0009387667 W = [0.09813061] b = [0.29937574] loss = 0.0009382542 W = [0.09850783] b = [0.2993767] loss = 0.00093801465 W = [0.09876581] b = [0.29937735] loss = 0.0009379023 W = [0.09894225] b = [0.2993778] loss = 0.00093785016 W = [0.09906292] b = [0.2993781] loss = 0.00093782565 W = [0.09914544] b = [0.2993783] loss = 0.00093781407 W = [0.09920189] b = [0.29937845] loss = 0.0009378086 W = [0.09924049] b = [0.29937854] loss = 0.0009378061 W = [0.09926689] b = [0.2993786] loss = 0.00093780505 W = [0.09928494] b = [0.29937866] loss = 0.00093780446 W = [0.09929729] b = [0.2993787] loss = 0.0009378045
三、把结果图显示出来
运行后
plt.scatter(x_data,y_data,c='r')
plt.plot(x_data,sess.run(W)*x_data+sess.run(b))
plt.show()
运行结果: