我就廢話不多說了，直接上代碼吧！

tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy)

TensorFlow經過使用梯度下降法對損失函數中的變量進行修改值，默認修改tf.Variable(tf.zeros([784,10]))

為Variable的參數。

train_step = tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy,var_list=[w,b])

也可以使用var_list參數來定義更新那些參數的值

#導入Minst數據集import input_datamnist = input_data.read_data_sets("data",one_hot=True) #導入tensorflow庫import tensorflow as tf #輸入變量，把28*28的圖片變成一維數組（丟失結構信息）x = tf.placeholder("float",[None,784]) #權重矩陣，把28*28=784的一維輸入，變成0-9這10個數字的輸出w = tf.Variable(tf.zeros([784,10]))#偏置b = tf.Variable(tf.zeros([10])) #核心運算，其實就是softmax（x*w+b）y = tf.nn.softmax(tf.matmul(x,w) + b) #這個是訓練集的正確結果y_ = tf.placeholder("float",[None,10]) #交叉熵，作為損失函數cross_entropy = -tf.reduce_sum(y_ * tf.log(y)) #梯度下降算法，最小化交叉熵train_step = tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy) #初始化，在run之前必須進行的init = tf.initialize_all_variables()#創建session以便運算sess = tf.Session()sess.run(init) #迭代1000次for i in range(1000): #獲取訓練數據集的圖片輸入和正確表示數字 batch_xs, batch_ys = mnist.train.next_batch(100) #運行剛才建立的梯度下降算法，x賦值為圖片輸入，y_賦值為正確的表示數字 sess.run(train_step,feed_dict = {x:batch_xs, y_: batch_ys}) #tf.argmax獲取最大值的索引。比較運算后的結果和本身結果是否相同。#這步的結果應該是[1,1,1,1,1,1,1,1,0,1...........1,1,0,1]這種形式。#1代表正確，0代表錯誤correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1)) #tf.cast先將數據轉換成float，防止求平均不準確。#tf.reduce_mean由于只有一個參數，就是上面那個數組的平均值。accuracy = tf.reduce_mean(tf.cast(correct_prediction,"float"))#輸出print(sess.run(accuracy,feed_dict={x:mnist.test.images,y_: mnist.test.labels}))

計算結果如下

"C:/Program Files/Anaconda3/python.exe" D:/pycharmprogram/tensorflow_learn/softmax_learn/softmax_learn.pyExtracting data/train-images-idx3-ubyte.gzExtracting data/train-labels-idx1-ubyte.gzExtracting data/t10k-images-idx3-ubyte.gzExtracting data/t10k-labels-idx1-ubyte.gzWARNING:tensorflow:From C:/Program Files/Anaconda3/lib/site-packages/tensorflow/python/util/tf_should_use.py:175: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02.Instructions for updating:Use `tf.global_variables_initializer` instead.2018-05-14 15:49:45.866600: W C:/tf_jenkins/home/workspace/rel-win/M/windows/PY/35/tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.2018-05-14 15:49:45.866600: W C:/tf_jenkins/home/workspace/rel-win/M/windows/PY/35/tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.0.9163 Process finished with exit code 0