TF——深度学习逻辑回归源码（加各种用到算法的解析）

解析在注释里：

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import os
os.environ["CUDA_VISIBLE_DEVICES"]="0"
mnist = input_data.read_data_sets("./data/MNIST_data/", one_hot=True)
learning_rate=0.01
training_epochs=25
batch_size=100  #一次训练所选取的样本数。
display_step=1

if __name__ =='__main__':
    x=tf.placeholder(tf.float32,[None,784])
    y=tf.placeholder(tf.float32,[None,10])

    print("PPPPPP")
    batch_xs, batch_ys = mnist.train.next_batch(batch_size)
    print(batch_xs.shape)
    print("LLLLLLLLLLLL")
    print(batch_ys.shape)

    '''Variable 初始化'''
    W=tf.Variable(tf.zeros([784,10]))
    b=tf.Variable(tf.zeros([10]))

    '''逻辑回归模型'''
    pred=tf.nn.softmax(tf.matmul(x,W)+b)

    '''构造代价函数'''
    '''
    tf.reduce_mean 函数用于计算张量tensor沿着指定的数轴（tensor的某一维度）上的的平均值，主要用作降维或者计算tensor（图像）的平均值。
    reduce_mean(input_tensor,
                    输入的待降维的tensor
                    axis=None,
                    指定的轴，如果不指定，则计算所有元素的均值
                    keep_dims=False,
                    是否降维度
                    name=None,
                    操作的名称
                    reduction_indices=None
                    用来指定轴
                    )
    '''
    cost = tf.reduce_mean(-tf.reduce_sum(y * tf.log(pred), reduction_indices=1))

    '''梯度下降最小值'''
    optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

    init = tf.global_variables_initializer()
    with tf.Session() as sess:
        sess.run(init)

        '''训练模型'''
        for epoch in range(training_epochs):
            avg_cost = 0
        total_batch = int(mnist.train.num_examples / batch_size)
        for i in range(total_batch):
            batch_xs, batch_ys = mnist.train.next_batch(batch_size)
            _, c = sess.run([optimizer, cost], feed_dict={x: batch_xs, y: batch_ys})
            # Conpute average loss
            avg_cost += c / total_batch
        if (epoch + 1) % display_step == 0:
            print("第:", '%04d步' % (epoch + 1), "标准差:", "{:.09f}".format(avg_cost))


        '''测试模型'''
        '''
        argmax(f(x))是使得 f(x)取得最大值所对应的变量点x(或x的集合)
        test = np.array([
            [1, 2, 3],
             [2, 3, 4], 
             [5, 4, 3], 
             [8, 7, 2]])
            np.argmax(test, 0)　　　＃输出：array([3, 3, 1]
            np.argmax(test, 1)　　　＃输出：array([2, 2, 0, 0]

            axis = 0:
            　　axis=0时比较每一列的元素，将每一列最大元素所在的索引记录下来，最后输出每一列最大元素所在的索引数组。
            axis = 1:
    　　axis=1的时候，将每一行最大元素所在的索引记录下来，最后返回每一行最大元素所在的索引数组。
        '''
        correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
        accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
        print("准确的:", accuracy.eval({x: mnist.test.images[:3000], y: mnist.test.labels[:3000]}))