tensforlow
标签: 机器学习 TensorFlow
[TOC]
文章主要讲述tensorflow中几个简单的模型
0. Tensorflow Bases
- datetype with tensorflow
| 名称 | 类型 | 说明 |
|---|---|---|
| constant | 常量，定义时给出 | tf.constant(2., tf.float32) |
| placeholder | 占位符，一般用于输入输出，运算前给出，也是程序参数输入的入口 | x = tf.placeholder(tf.float32) |
| Variable | 变量，可以在运算过程中访问（赋值或者读取） | W = tf.Variable([.3], tf.float32) |
1. 线性回归（linear regression）
linear regression algorithm helps to predict scores ont the variable Y from the scores on the variable X.
linear regression theoty
simple example
# coding:utf-8
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
# parameters
learning_rate = 0.001
training_step = 1000
display_step = 50
# training data
x_train = np.linspace(-1, 1, 101).astype('float32')
y_train = 2 * x_train + np.random.randn(len(x_train)).astype('float32') * 0.33
train_X = np.asarray([3.3, 4.4, 5.5, 6.71, 6.93, 4.168, 9.779, 6.182, 7.59,
2.167, 7.042, 10.791, 5.313, 7.997, 5.654, 9.27, 3.1])
train_Y = np.asarray([1.7, 2.76, 2.09, 3.19, 1.694, 1.573, 3.366, 2.596, 2.53,
1.221, 2.827, 3.465, 1.65, 2.904, 2.42, 2.94, 1.3])
# define model y = W*x+b
W = tf.Variable(tf.zeros([1.0]), tf.float32)
b = tf.Variable(tf.zeros([1.0]), tf.float32)
x = tf.placeholder(tf.float32)
y = tf.placeholder(tf.float32)
linear_model = W * x + b
# cost function
cost = tf.square(linear_model - y)
# training algorythom
train_op = tf.train.GradientDescentOptimizer(0.001).minimize(cost)
# initalize variables
init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init)
# training
for i in range(100):
sess.run(train_op, feed_dict={x: train_X, y: train_Y})
plt.figure('Figure1')
plt.scatter(train_X,train_Y, c='b', marker='o')
plt.plot(train_X, sess.run(W) * train_X + sess.run(b))
plt.show()
print (sess.run(W), sess.run(b))#logical regression