# -*- coding: utf-8 -*-
"""
Created on Wed Feb 28 23:06:31 2018
@author: xkk
"""
import tensorflow as tf
import numpy as np
x_data=np.random.rand(100).astype(np.float32)
z_data=np.random.rand(100).astype(np.float32)
y_data=x_data*2+z_data*4+7
alpha = 0.01
step = 1000
Weights1 = tf.Variable(tf.random_uniform([1],-1,1))
Weights2 = tf.Variable(tf.random_uniform([1],-1,1))
biases = tf.Variable(tf.zeros(1))
y=x_data*Weights1+z_data*Weights2+biases
loss = tf.reduce_mean(tf.square(y-y_data))
optimizer = tf.train.GradientDescentOptimizer(alpha)
train = optimizer.minimize(loss)
init = tf.initialize_all_variables()
sess = tf.Session()
sess.run(init)
for i in range(step):
sess.run(train)
if i%20 == 0:
print(sess.run(Weights1),sess.run(Weights2),sess.run(biases))
Python_tensorflow实战1.多元回归
最新推荐文章于 2022-12-29 14:15:19 发布
这篇博客通过Python和TensorFlow库演示了如何进行多元线性回归。作者首先生成了随机数据,然后定义了权重和偏置变量,构建了损失函数,并使用梯度下降优化器进行训练。经过1000次迭代,模型逐渐学习到输入数据的线性关系。
摘要由CSDN通过智能技术生成