機器學習進階筆記之一 | TensorFlow安裝與入門

引言

TensorFlow是Google基於DistBelief進行研發的第二代人工智能學習系統，被廣泛用於語音識別或圖像識別等多項機器深度學習領域。其命名來源於本身的運行原理。Tensor（張量）意味着N維數組，Flow（流）意味着基於數據流圖的計算，TensorFlow代表着張量從圖象的一端流動到另一端計算過程，是將復雜的數據結構傳輸至人工智能神經網中進行分析和處理的過程。

TensorFlow完全開源，任何人都可以使用。可在小到一部智能手機、大到數千台數據中心服務器的各種設備上運行。

『機器學習進階筆記』系列是將深入解析TensorFlow系統的技術實踐，從零開始，由淺入深，與大家一起走上機器學習的進階之路。

CUDA與TensorFlow安裝

按以往經驗，TensorFlow安裝一條pip命令就可以解決，前提是有fq工具，沒有的話去找找牆內別人分享的地址。而坑多在安裝支持gpu，需預先安裝英偉達的cuda，這里坑比較多，推薦使用ubuntu deb的安裝方式來安裝cuda，run.sh的方式總感覺有很多問題，cuda的安裝具體可以參考。注意鏈接里面的tensorflow版本是以前的，tensorflow 現在官方上的要求是cuda7.5+cudnnV4，請在安裝的時候注意下。

Hello World

 import tensorflow as tf hello = tf.constant('Hello, TensorFlow!') sess = tf.Session() print sess.run(hello)

首先，通過tf.constant創建一個常量，然后啟動Tensorflow的Session，調用sess的run方法來啟動整個graph。
接下來我們做下簡單的數學的方法：

 import tensorflow as tf a = tf.constant(2) b = tf.constant(3) with tf.Session() as sess: print "a=2, b=3" print "Addition with constants: %i" % sess.run(a+b) print "Multiplication with constants: %i" % sess.run(a*b) # output a=2, b=3 Addition with constants: 5 Multiplication with constants: 6

接下來用tensorflow的placeholder來定義變量做類似計算：
placeholder的使用見 https://www.tensorflow.org/versions/r0.8/api_docs/python/io_ops.html#placeholder

 import tensorflow as tf
 a = tf.placeholder(tf.int16)
 b = tf.placeholder(tf.int16)
 add = tf.add(a, b)
 mul = tf.mul(a, b)
 with tf.Session() as sess:
     # Run every operation with variable input
     print "Addition with variables: %i" % sess.run(add, feed_dict={a: 2, b: 3}) print "Multiplication with variables: %i" % sess.run(mul, feed_dict={a: 2, b: 3}) # output: Addition with variables: 5 Multiplication with variables: 6 matrix1 = tf.constant([[3., 3.]]) matrix2 = tf.constant([[2.],[2.]]) with tf.Session() as sess: result = sess.run(product) print result

線性回歸

以下代碼來自GitHub - aymericdamien/TensorFlow-Examples: TensorFlow Tutorial and Examples for beginners,僅作學習用

 import tensorflow as tf import numpy import matplotlib.pyplot as plt rng = numpy.random # Parameters learning_rate = 0.01 training_epochs = 2000 display_step = 50 # Training Data train_X = numpy.asarray([3.3,4.4,5.5,6.71,6.93,4.168,9.779,6.182,7.59,2.167,7.042,10.791,5.313,7.997,5.654,9.27,3.1]) train_Y = numpy.asarray([1.7,2.76,2.09,3.19,1.694,1.573,3.366,2.596,2.53,1.221,2.827,3.465,1.65,2.904,2.42,2.94,1.3]) n_samples = train_X.shape[0] # tf Graph Input X = tf.placeholder("float") Y = tf.placeholder("float") # Create Model # Set model weights W = tf.Variable(rng.randn(), name="weight") b = tf.Variable(rng.randn(), name="bias") # Construct a linear model activation = tf.add(tf.mul(X, W), b) # Minimize the squared errors cost = tf.reduce_sum(tf.pow(activation-Y, 2))/(2*n_samples) #L2 loss optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost) #Gradient descent # Initializing the variables init = tf.initialize_all_variables() # Launch the graph with tf.Session() as sess: sess.run(init) # Fit all training data for epoch in range(training_epochs): for (x, y) in zip(train_X, train_Y): sess.run(optimizer, feed_dict={X: x, Y: y}) #Display logs per epoch step if epoch % display_step == 0: print "Epoch:", '%04d' % (epoch+1), "cost=", \ "{:.9f}".format(sess.run(cost, feed_dict={X: train_X, Y:train_Y})), \ "W=", sess.run(W), "b=", sess.run(b) print "Optimization Finished!" print "cost=", sess.run(cost, feed_dict={X: train_X, Y: train_Y}), \ "W=", sess.run(W), "b=", sess.run(b) #Graphic display plt.plot(train_X, train_Y, 'ro', label='Original data') plt.plot(train_X, sess.run(W) * train_X + sess.run(b), label='Fitted line') plt.legend() plt.show()

邏輯回歸

 import tensorflow as tf # Import MINST data from tensorflow.examples.tutorials.mnist import input_data mnist = input_data.read_data_sets("/tmp/data/", one_hot=True) # Parameters learning_rate = 0.01 training_epochs = 25 batch_size = 100 display_step = 1 # tf Graph Input x = tf.placeholder(tf.float32, [None, 784]) # mnist data image of shape 28*28=784 y = tf.placeholder(tf.float32, [None, 10]) # 0-9 digits recognition => 10 classes # Set model weights W = tf.Variable(tf.zeros([784, 10])) b = tf.Variable(tf.zeros([10])) # Construct model pred = tf.nn.softmax(tf.matmul(x, W) + b) # Softmax # Minimize error using cross entropy cost = tf.reduce_mean(-tf.reduce_sum(y*tf.log(pred), reduction_indices=1)) # Gradient Descent optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost) # Initializing the variables init = tf.initialize_all_variables() # Launch the graph with tf.Session() as sess: sess.run(init) # Training cycle for epoch in range(training_epochs): avg_cost = 0. total_batch = int(mnist.train.num_examples/batch_size) # Loop over all batches for i in range(total_batch): batch_xs, batch_ys = mnist.train.next_batch(batch_size) # Run optimization op (backprop) and cost op (to get loss value) _, c = sess.run([optimizer, cost], feed_dict={x: batch_xs, y: batch_ys}) # Compute average loss avg_cost += c / total_batch # Display logs per epoch step if (epoch+1) % display_step == 0: print "Epoch:", '%04d' % (epoch+1), "cost=", "{:.9f}".format(avg_cost) print "Optimization Finished!" # Test model correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1)) # Calculate accuracy accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) print "Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}) # result : Epoch: 0001 cost= 29.860467369 Epoch: 0002 cost= 22.001451784 Epoch: 0003 cost= 21.019925554 Epoch: 0004 cost= 20.561320320 Epoch: 0005 cost= 20.109135756 Epoch: 0006 cost= 19.927862290 Epoch: 0007 cost= 19.548687116 Epoch: 0008 cost= 19.429119071 Epoch: 0009 cost= 19.397068211 Epoch: 0010 cost= 19.180813479 Epoch: 0011 cost= 19.026808132 Epoch: 0012 cost= 19.057875510 Epoch: 0013 cost= 19.009575057 Epoch: 0014 cost= 18.873240641 Epoch: 0015 cost= 18.718575359 Epoch: 0016 cost= 18.718761925 Epoch: 0017 cost= 18.673640560 Epoch: 0018 cost= 18.562128253 Epoch: 0019 cost= 18.458205289 Epoch: 0020 cost= 18.538211225 Epoch: 0021 cost= 18.443384213 Epoch: 0022 cost= 18.428727668 Epoch: 0023 cost= 18.304270616 Epoch: 0024 cost= 18.323529782 Epoch: 0025 cost= 18.247192113 Optimization Finished! (10000, 784) Accuracy 0.9206

這里有個小插曲，ipython notebook在一個notebook打開時，一直在占用GPU資源，可能是之前有一個notebook一直打開着，然后占用着GPU資源，然后在計算Accuracy的”InternalError: Dst tensor is not initialized.” 然后找了github上面也有這個問題InternalError: Dst tensor is not initialized.,可以肯定是GPU的memory相關的問題，所以就嘗試加上tf.device(‘/cpu:0’)，將Accuracy這步拉到cpu上計算，但是又出現OOM的問題，最后nvidia-smi時，發現有一個python腳本一直占用3g多的顯存，把它kill之后恢復了，之前還比較吐槽怎么可能10000*784個float就把顯存撐爆呢，原來是自己的問題。

這里邏輯回歸，model是一個softmax函數用來做多元分類，大概意思是選擇10當中最后預測概率最高作為最終的分類。

其實基本的tensorflow沒有特別好講的，語法的課程什么可以去看看基本的文檔，之后我會找一點經典有趣的tensorflow的代碼應用來看看，畢竟『show me the code 』才是程序猿應有的態度。

本文由『UCloud內核與虛擬化研發團隊』提供。

關於作者：

Burness（@段石石）， UCloud平台研發中心深度學習研發工程師，tflearn Contributor，做過電商推薦、精准化營銷相關算法工作，專注於分布式深度學習框架、計算機視覺算法研究，平時喜歡玩玩算法，研究研究開源的項目，偶爾也會去一些數據比賽打打醬油，生活中是個極客，對新技術、新技能痴迷。