基於tensorflow的簡單線性回歸模型
#!/usr/local/bin/python3
##ljj [1]
##linear regression model
import tensorflow as tf
import matplotlib.pyplot as plt
x_ = [11,14,22,29,32,40,44,55,59,60,69,77]
y_res = [123,135,155,167,177,189,200,240,250,255,277,298]
w = tf.Variable(tf.ones([1]),dtype="float32")
b = tf.Variable(tf.ones([1]),dtype="float32")
y_predict = tf.placeholder(tf.float32)
x = tf.placeholder(tf.float32)
with tf.Session() as sess:
y_predict = w*x+b
loss = tf.reduce_mean(tf.square(y_res-y_predict))
train = tf.train.AdamOptimizer(0.7).minimize(loss)
sess.run(tf.global_variables_initializer())
for i in range(len(x_)):
# train.run(feed_dict={x:x_[i], y_predict:y_res[i]})
w_,b_,_= sess.run([w,b,train],feed_dict={x:x_[i], y_predict:y_res[i]})
print(w_,b_)
plt.plot(x_,y_res,‘.‘)
plt.plot(x_,x_*w_+b_,‘-‘)
plt.show()
主機環境：MacbookPro，tensoflow版本1.4
輸出結果：
ljjdeMBP:linear_regression lingjiajun$ ./linear_regression.py
/usr/local/Cellar/python3/3.6.2/Frameworks/Python.framework/Versions/3.6/lib/python3.6/importlib/_bootstrap.py:205: RuntimeWarning: compiletime version 3.5 of module ‘tensorflow.python.framework.fast_tensor_util‘ does not match runtime version 3.6
return f(*args, **kwds)
2018-04-27 22:20:05.963003: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
[ 4.78998518] [ 5.67698431]
-------以上輸出分別是擬合出的Weight，Bias值。
基於tensorflow的簡單線性回歸模型