tensorflow示例代码注释1

#!/usr/bin/env python

import tensorflow as tf

//导入numpy科学计算包

import numpy as np

//numpy.linspace(start,
stop, num=50, endpoint=True, retstep=False,
dtype=None)，分配从-1到1中间的1０１个数，包含-1和1，构成一个线性数组trX

trX = np.linspace(-1, 1, 101)

//randn 返回一个样本，具有标准正态分布,ndarray.shape：数组的维度,numpy的array就是一个矩阵，不同于python中的list,tuple,dict,set

trY = 2 * trX + np.random.randn(*trX.shape) * 0.33 # create a y value which is approximately linear but with some random noise

//placeholder，占位符

X = tf.placeholder("float") # create symbolic variables

Y = tf.placeholder("float")

def model(X, w):

    return tf.mul(X, w) # lr is just X*w so this model line is pretty simple

w = tf.Variable(0.0, name="weights") # create a shared variable (like theano.shared) for the weight matrix

y_model = model(X, w)

//方差
cost = tf.square(Y - y_model) # use square error for cost function

//梯度下降类，最小目标cost,学习率0.01

train_op = tf.train.GradientDescentOptimizer(0.01).minimize(cost) # construct an optimizer to minimize cost and fit line to my data

# Launch the graph in a session

with tf.Session() as sess:

    # you need to initialize variables (in this case just variable W)

    tf.initialize_all_variables().run()

    for i in range(100):

//python的zip函数，对矩阵求T，如([1,2],[3,4])---->([1,3],[2,4])

        for (x, y) in zip(trX, trY):

            sess.run(train_op, feed_dict={X: x, Y: y})

    print(sess.run(w))  # It should be something around 2