# 在用tensorflow前先打開
$ source ~/tensorflow/bin/activate
# 激活後會有標識
(tensorflow)$
#用完後關閉
(tensorflow)$ deactivate
# 刪除tensorflow
$ rm -r ~/tensorflow
1. Getting Started With Tensorflowhtml
Tensorflow的基本單元是張量(tensor)，張量的階(rank)是它的維度(the number of dimension)。node
Tensorflow由一系列運算圖(computational graph)組成。點有多種類型，包括常數(constant)、變量(variable)。python
要操做computational graph中的node，要用到session。git
sess = tf.Session() print(sess.run([node1, node2]))
可視化： TensorBoard能夠用來表現computational graph。github
用placeholder能夠完成相似函數(function)的功能數組
a = tf.placeholder(tf.float32) b = tf.placeholder(tf.float32) adder_node = a + b # + provides a shortcut for tf.add(a, b)
print(sess.run(adder_node, {a: 3, b: 4.5}))
利用變量(variables)給節點添加參數網絡
2. Deep MNIST for Expertssession
InteractiveSession class is more convenient and flexible.ide
import tensorflow as tf sess = tf.InteractiveSession()
Placeholders with shape argument can automatically catch bugs regarding on manipulation on inconsistent tensor shape.函數
x = tf.placeholder(tf.float32, shape=[None, 784]) # None means that the number of sample is variable and the length of each sample is 784 (28 * 28)
包含Deep convolutional neural networks的使用教程。
3. Read Data
batchsize: 每次訓練在訓練集中取得樣本數
iteration: 使用batchsize個樣本訓練一次
epoch: 使用訓練集中的所有樣本訓練一次
# 讀入的數據以列表存儲 filename = [] # tf.train.string_input_producer() 產生一個文件名隊列 filename_queue = tf.train.string_input_producer(filename, shuffle=False, num_epochs=5）# reader從文件名隊列中讀數據。對應的方法是reader.read reader = tf.WholeFileReader() key, value = reader.read(filename_queue) # tf.train.string_input_producer定義了一個epoch變量，要對它進行初始化 tf.local_variables_initializer().run() # 使用start_queue_runners以後，纔會開始填充隊列 threads = tf.train.start_queue_runners(sess=sess)
詳見: http://blog.csdn.net/buptgshengod/article/details/72956846
numpy
import numpy as np np.loadtext(c) # return ndarray
np.ones() # return a new array of given shape and type, filled with ones
i = 1 * 10 ** (-10) # ** means power, and its priority is higher than *
np.eye(3, dtype = int) # diagonal matrix, consist of 0 and 1
vector = np.hstack((vector, temp)) # concatenation along the second axis (horizontally in 2 dimension)
np.save(filename, array, fmt = '%d') # save the data of int format
4. make hand dirty
numpy一位數組轉置只須要操做shape
array_1d = np.array([1, 2]) array_1d.shape = (2, 1) # finished array_2d.transpose() # 二維數組的轉置方法
python assert斷言是聲明其布爾值必須爲真的斷定
assert len(lists) >=5 # 若爲假則觸發異常
使用名稱前的單下劃線，用於指定該名稱屬性爲「私有」，其只供內部使用。 （_spam)
collections是一個集合模塊，提供許多有用的集合類，如namedtuple(提供一個列表的集合類）
>>> from collections import namedtuple >>> Point = namedtuple('Point', ['x', 'y']) >>> p = Point(1, 2) >>> p.x 1 >>> p.y 2
5. Model
strides 中的四個參數，第一個是 the number of images，第二個是 the height of images，第三個是 the width of images，第四個是 the number of channel
小白學Tensorflow之卷積神經網絡 https://www.jianshu.com/p/70c8d6663b00
深刻MNIST http://www.tensorfly.cn/tfdoc/tutorials/mnist_pros.html
Batch Normalization 批標準化 https://morvanzhou.github.io/tutorials/machine-learning/tensorflow/5-13-BN/
martin-gorner/tensorflow-mnist-tutorial (batch normalization)https://github.com/martin-gorner/tensorflow-mnist-tutorial
6. 求AUC
import tensorflow as tf a = tf.Variable([0.1, 0.5]) b = tf.Variable([0.2, 0.6]) auc = tf.contrib.metrics.streaming_auc(a, b) sess = tf.Session() sess.run(tf.initialize_all_variables()) sess.run(tf.initialize_local_variables()) # try commenting this line and you'll get the error train_auc = sess.run(auc) print(train_auc)
https://stackoverflow.com/questions/39435341/how-to-calculate-auc-with-tensorflow
7. 可視化 （好）
Tensorflow的可視化工具Tensorboard的初步使用
https://blog.csdn.net/mao_feng/article/details/54731098
tensorflow + mnist + cnn + batch normalization
https://github.com/hwalsuklee/tensorflow-mnist-cnn
8. 過擬合狀況及解決
dropout https://blog.csdn.net/smf0504/article/details/55254818
batch normalization https://blog.csdn.net/whitesilence/article/details/75667002
slim cnn 代碼 https://www.2cto.com/kf/201706/649266.html
9.論文寫法
https://blog.csdn.net/fengbingchun/article/details/50529500 cnn 結構