使用tensorflow进行简单的线性回归
标签(空格分隔): tensorflow
数据准备
- 使用np.random.uniform()生成x方向的数据
- 使用np.random.uniform()生成bias数据
- 直线方程为y=0.1x + 0.2
- 使用梯度下降算法
代码
import numpy as npimport tensorflow as tfpath = 'D:\tensorflow_quant\ailib\log_tmp'# 生成x数据points = 100vectors = []for i in range(points): # y=0.1*x + 0.2 x = np.random.uniform(0, 0.66) y = x * 0.1 + 0.2 + np.random.uniform(0, 0.04) vectors.append([x, y])x_data = [v[0] for v in vectors]y_data = [v[1] for v in vectors]#形成计算图w = tf.Variable(tf.random_uniform([1], -1.0, 1.0))b = tf.Variable(tf.zeros([1]))y = w * x_data + b#定义损失函数loss = tf.reduce_mean(tf.square(y-y_data))#定义优化器optimizer = tf.train.GradientDescentOptimizer(0.5)train = optimizer.minimize(loss)#对计算图开始计算with tf.Session() as sess: init = tf.global_variables_initializer() sess.run(init) for step in range(1000): sess.run(train) if step%5==0: print(step,sess.run(loss),sess.run(w),sess.run(b)) #生成计算日志 writer = tf.Summary.FileWriter(path,sess.graph)