教程的第一部分主要讲的是mnist_softmax.py里面的代码:
首先是下载和读取mnist数据集,两行代码来实现:
from tensorflow.examples.tutorials.mnist import input_data mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)A one-hot vector is a vector which is 0 in most dimensions, and 1 in asingle dimension.
第一行是导入input_data模块,该模块的代码:
import tensorflow as tf from tensorflow.contrib.learn.python.learn.datasets.mnist import read_data_sets
发现其实没有直接的input_data.read_data_sets()函数,而是又导入了另一个模块,该模块名叫read_data_sets,找到这个文件,里面的代码是:
def read_data_sets(train_dir, fake_data=False, one_hot=False, dtype=dtypes.float32, reshape=True, validation_size=5000): if fake_data: def fake(): return DataSet([], [], fake_data=True, one_hot=one_hot, dtype=dtype) train = fake() validation = fake() test = fake() return base.Datasets(train=train, validation=validation, test=test) TRAIN_IMAGES = 'train-images-idx3-ubyte.gz' TRAIN_LABELS = 'train-labels-idx1-ubyte.gz' TEST_IMAGES = 't10k-images-idx3-ubyte.gz' TEST_LABELS = 't10k-labels-idx1-ubyte.gz' local_file = base.maybe_download(TRAIN_IMAGES, train_dir, SOURCE_URL + TRAIN_IMAGES) with open(local_file, 'rb') as f: train_images = extract_images(f) local_file = base.maybe_download(TRAIN_LABELS, train_dir, SOURCE_URL + TRAIN_LABELS) with open(local_file, 'rb') as f: train_labels = extract_labels(f, one_hot=one_hot) local_file = base.maybe_download(TEST_IMAGES, train_dir, SOURCE_URL + TEST_IMAGES) with open(local_file, 'rb') as f: test_images = extract_images(f) local_file = base.maybe_download(TEST_LABELS, train_dir, SOURCE_URL + TEST_LABELS) with open(local_file, 'rb') as f: test_labels = extract_labels(f, one_hot=one_hot) if not 0 <= validation_size <= len(train_images): raise ValueError( 'Validation size should be between 0 and {}. Received: {}.' .format(len(train_images), validation_size)) validation_images = train_images[:validation_size] validation_labels = train_labels[:validation_size] train_images = train_images[validation_size:] train_labels = train_labels[validation_size:] train = DataSet(train_images, train_labels, dtype=dtype, reshape=reshape) validation = DataSet(validation_images, validation_labels, dtype=dtype, reshape=reshape) test = DataSet(test_images, test_labels, dtype=dtype, reshape=reshape) return base.Datasets(train=train, validation=validation, test=test)这段代码,主要就是来准备训练集、测试集、验证集。其中下载下来的训练集前部分变成了准备好的训练集,后一部分变成了验证集。
然后将这个28*28的图像变为1*784的向量
最终的结果是把55000张图像,变成一个tensor张量[55000,784]。每张图像对应一个one-hotx向量1*10,所以label变成tensor[55000,10]
接下来是Softmax Regressions介绍:
其实解决的是10分类的问题,主要分为两步:首先是得出输入属于某一class的evidence,然后,将这个evidence转化成概率。
这里的evidence其实类似于加权和:
然后就是得到概率值y:
y = softmax(evidence),下面的是工作流程:
然后就是通过softmax()获得概率值:
以矩阵的形式表示:
最终的简洁表示为:
y = softmax(Wx+b)
使用tensorflow完成回归(implementing the regression)
首先导入tensorflow
import tensorflow as tf
创建x用来存放图片数据
x = tf.placeholder(tf.float32, [None, 784])
然后创建两个变量,存放权重参数与偏移参数
W = tf.Variable(tf.zeros([784, 10])) b = tf.Variable(tf.zeros([10]))
然后使用一行代码完成分类任务。
y = tf.nn.softmax(tf.matmul(x, W) + b)
训练(training):
训练的过程本质上是使LOSS变小的过程,若LOSS接近0,则说明训练的效果好。
交叉熵是其中一个方法来定义LOSS模型。
在使用交叉熵之前,需要将真值(金标准)(标签)导入进来:
y_ = tf.placeholder(tf.float32, [None, 10])所以:
cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y), reduction_indices=[1]))
计算出来LOSS之后,就要利用LOSS反向传播,使用的方法为梯度下降法。(还有很多其他的优化算法)
train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)
在训练之前,需要将所有的变量初始化,
init = tf.initialize_all_variables()
现在可以在会话层中初始化了(Session)
sess = tf.Session() sess.run(init)
训练1000次的Python代码如下:
for i in range(1000): batch_xs, batch_ys = mnist.train.next_batch(100) sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys})
最后是测试,输出精度:
correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) print(sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels}))
完成任务。我是在ipython notebook中跑的例子。说一下这个过程中遇到的一些问题。
1、一开始我用的是下载的tensorflow 里面的mnist_softmax.py里面的代码,实在是坑,因为里面没有定义Session会话,所以会报错,根据官方教程把里面的Session加上去。
2、再运行,没有报错,但是也没有输出结果,很纳闷,然后我在终端里面输入:python 路径(mnist_softmax.py的路径)竟然有结果。
3、后来发现是因为代码里面定义了一个main()函数,所以只会编译,没有运行。
4、然后我在另一个cell里面调用这个函数,报错!关于input_data的问题,发现是因为找不到数据的位置,所以给它指定了位置,运行成功,下面附上完整代码与结果,(总结就是跑程序的时候,别用下载下来的代码,用官方教程一步一步写的代码)
from __future__ import absolute_import from __future__ import division from __future__ import print_function import argparse import tensorflow as tf from tensorflow.examples.tutorials.mnist import input_data def main(_): print("测试") mnist = input_data.read_data_sets("../../MNIST_data/",one_hot = True) x = tf.placeholder(tf.float32,[None,784]) W = tf.Variable(tf.zeros([784,10])) b = tf.Variable(tf.zeros(10)) y = tf.matmul(x,W) + b y_ = tf.placeholder(tf.float32,[None,10]) cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(y,y_)) train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy) init = tf.initialize_all_variables() sess = tf.Session() sess.run(init) for i in range(1000): batch_xs, batch_ys = mnist.train.next_batch(100) sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys}) correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(y_,1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32)) print(sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels})) main(_)
测试结果:
测试 Extracting ../../MNIST_data/train-images-idx3-ubyte.gz Extracting ../../MNIST_data/train-labels-idx1-ubyte.gz Extracting ../../MNIST_data/t10k-images-idx3-ubyte.gz Extracting ../../MNIST_data/t10k-labels-idx1-ubyte.gz 0.9142
正确率91%.
备注:运行环境:ubuntu14.04+python2.7+ipython notebook环境
参考:https://www.tensorflow.org/versions/r0.11/tutorials/mnist/beginners/index.html