ubuntu16.04 ananconda tensorflow
ABSTRACT
descripe :
date : 2018.06.06
platform : NVIDIA GeForce 1080Ti
version : ubuntu - 16.04 LTS(自帶的是python3.5，所以在安裝tensorflow時，需要注意版本)
NVIDIA-driver - 384.13
CUDA - 9.0
cuDNN - 7.1.2
tensorflow - 1.8
==========================
首先在search your computer中輸入upd，進入software&updates，更改下載源（推薦清華或阿里作為下載源。）
首先sudo apt update，然後sudo apt upgrade
==========================
STEP 1 : install nvidia driver
reference : https://blog.csdn.net/u012759136/article/details/53355781
修改驅動檔案許可權
sudo chmod 777 NVIDIA-Linux-x86_64-410.78.run
1. sudo apt-get remove –purge nvidia*
2. sudo vim /etc/modprobe.d/blacklist.conf (sudo apt-get install vim-gtk)
3. add "blacklist nouveau" at the end of file
4. sudo update-initramfs -u
5. sudo reboot
6. Ctrl+Alt+F1
7. lsmod | grep nouveau
8. sudo /etc/init.d/lightdm stop
9. cd ${nvidia_driver_path}
10. sudo sh NVIDIA-Linux-x86_64-384.130.run –no-x-check –no-nouveau-check –no-opengl-files
安裝(注意 引數)
sudo ./NVIDIA-Linux-x86_64-384.13.run –no-opengl-files
–no-opengl-files 只安裝驅動檔案，不安裝OpenGL檔案。這個引數最重要
–no-x-check 安裝驅動時不檢查X服務
–no-nouveau-check 安裝驅動時不檢查nouveau
後面兩個引數可不加。
在安裝驅動的時候，有一步問你”Would you like to run the nvidia-xconfig utility to automatically update your X configuration file…”什麼的，選擇 No
11. sudo reboot
Please run the commands, if the desktop can not login after reboot.
1. cd /usr/share/X11/xorg.conf.d/
2. sudo mv nvidia-drm-outputclass.conf nvidia-drm-outputclass.conf.bak
3. sudo reboot
==========================
STEP 2: install CUDA
1. Download CUDA 9.0 at https://developer.nvidia.com/cuda-90-download-archive
select Linux, x86_64, Ubuntu, 16.04, runfile(local) and download base installer,(the download file name is "cuda_9.0.176_384.81_linux.run")
2. cd ${cuda_9.0_path}
3. sudo sh cuda_9.0.176_384.81_linux.run
4. enter n, y, '\n', y, y, '\n' step by step, like this. ('\n' means enter)
有一個關鍵是會讓你選擇是否安裝Graphics Driver for Linux-x86_64：XXXX版本，這個地方必須選擇no！否則會覆蓋之前的驅動。
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 384.81?
(y)es/(n)o/(q)uit: n
Install the CUDA 9.0 Toolkit?
(y)es/(n)o/(q)uit: y
Enter Toolkit Location
[ default is /usr/local/cuda-9.0 ]:
Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y
Install the CUDA 9.0 Samples?
(y)es/(n)o/(q)uit: y
Enter CUDA Samples Location
[ default is /home/ai_group ]:
Installing the CUDA Toolkit in /usr/local/cuda-9.0 ...
Missing recommended library: libGLU.so
Missing recommended library: libX11.so
Missing recommended library: libXi.so
Missing recommended library: libXmu.so
Missing recommended library: libGL.so
Installing the CUDA Samples in /home/ai_group ...
Copying samples to /home/ai_group/NVIDIA_CUDA-9.0_Samples now...
Finished copying samples.
===========
= Summary =
===========
Driver: Not Selected
Toolkit: Installed in /usr/local/cuda-9.0
Samples: Installed in /home/ai_group, but missing recommended libraries
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least 384.00 is required for CUDA 9.0 functionality to work.
To install the driver using this installer, run the following command, replacing <CudaInstaller> with the name of this run file:
sudo <CudaInstaller>.run -silent -driver
Logfile is /tmp/cuda_install_3795.log
[email protected]:~/Downloads/install required package$ sudo sh cuda_9.0.176_384.81_linux.run -silent -driver
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~``
5.1 gedit ./.bashrc
install CUDA 的 5. gedit ./.bashrc，請注意bashrc檔案在home路徑下，所以需要進入home路徑，開啟bashrc檔案進行編輯。
5.2 source ~/.bashrc
5.3 關閉終端，然後再重啟一個新的終端，這一步很重要
6. add environment path at the end
export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64$:LD_LIBRARY_PATH
==========================
STEP 3: install cuDNN
1. Download cuDNN for CUDA 9.0 at https://developer.nvidia.com/rdp/cudnn-archive
2. Extract the cuDNN
3. cd ${cuDNN_extract_folder_path}
4. sudo cp ./include/cudnn.h /usr/local/cuda/include
5. sudo cp ./lib64/libcudnn* /usr/local/cuda/lib64
6. sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*
==========================
STEP 4: install Anaconda (我安裝的對應於python3.6，注意tensorflow對應python的版本)
1、先從官網https://www.anaconda.com/download/#linux或者https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/下載對應的anaconda版本
2、下載完成後進入下載目錄執行以下語句
$ bash ./Anaconda3-5.2.0-Linux-x86_64.sh
#此處將檔名替換成你自己的檔名
3、接下來的安裝提示按回車進入下一步，按q跳過License文件，最後輸入yes確認
接下來讓我們輸入安裝路徑，沒特殊情況直接回車預設路徑就好
4、安裝完成後程式提示我們是否把Anaconda3的binary路徑加入到.bashrc，建議新增，這樣以後python和ipython命令就會自動使用anaconda Python3.6環境了（全程Yes，預設安裝在當前使用者的home目錄下，注意，最後一步，是問你要不要新增環境變數，預設是ＮＯ，最好選擇YES.）
5、可以在終端輸入$ conda list檢視是否安裝成功
=========================
STEP 5: 使用anaconda安裝tensorflow
1、建立conda計算環境
conda create -n tensorflow python=3.6
過程中輸入y繼續進行
2、啟用環境,使用 conda 安裝 TensorFlow
source activate tensorflow
3、安裝tensorflow
pip install --ignore-installed --upgrade https://mirrors.tuna.tsinghua.edu.cn/tensorflow/linux/gpu/tensorflow_gpu-1.5.0rc1-cp36-cp36m-linux_x86_64.whl
4、測試tensorflow
按照圖中命令輸入
=========================
STEP 6: pycharm安裝
1、pycharm下載
https://www.jetbrains.com/pycharm/download/#section=linux
社群版對於我們這些學生大體夠用，當然你也可以下載社群版，然後用學校郵箱進行註冊，可免費使用
2、解壓下載檔案
進入下載目錄，執行以下解壓命令
tar -xzvf pycharm-professional-2018.1.tar.gz
解壓後，會生成一個檔案
這裡寫圖片描述
3、進入檔案bin目錄，進行安裝
cd pycharm-2018.1/bin
sh ./pycharm.sh
5、pycharm匯入tensorflow
File->Settings->Project: 專案名->Project Interpreter
點選右上角齒輪圖形
選擇Conda Enviroment,再選擇Existing enviroment
選擇你的tensorflow下的python3.6,例如我的:/home/lee/anaconda3/envs/tensorflow/bin/python3.6
——————————————————————————————————
測試是否安裝成功
import tensorflow as tf
hello=tf.constant（'Hello, Tensorflow'）
sess = tf.Session()
print(sess.run(hello))
=========================
STEP 7: tensorflow環境下安裝python包
1、開啟終端，啟用tensorflow
source activate tensorflow
2、繼續安裝numpy、pandas、matplotlib，安裝方法很簡單
conda install numpy
過程中選擇yes
=============================================
STEP 8:https://blog.csdn.net/weixin_40362972/article/details/79838662
八、mnist測試
1、下載mnist資料集
進入http://yann.lecun.com/exdb/mnist/，下載如下圖四個檔案
這裡寫圖片描述
注意不要去解壓下載過來的檔案，然後將檔案複製到你的python的Project的目錄下
在pycharm輸入以下程式碼進行測試
# load MNIST data
import tensorflow.examples.tutorials.mnist.input_data as input_data
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
# start tensorflow interactiveSession
import tensorflow as tf
sess = tf.InteractiveSession()
# weight initialization
def weight_variable(shape):
initial = tf.truncated_normal(shape, stddev=0.1)
return tf.Variable(initial)
def bias_variable(shape):
initial = tf.constant(0.1, shape=shape)
return tf.Variable(initial)
# convolution
def conv2d(x, W):
return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')
# pooling
def max_pool_2x2(x):
return tf.nn.max_pool(x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
# Create the model
# placeholder
x = tf.placeholder("float", [None, 784])
y_ = tf.placeholder("float", [None, 10])
# variables
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))
y = tf.nn.softmax(tf.matmul(x, W) + b)
# first convolutinal layer
w_conv1 = weight_variable([5, 5, 1, 32])
b_conv1 = bias_variable([32])
x_image = tf.reshape(x, [-1, 28, 28, 1])
h_conv1 = tf.nn.relu(conv2d(x_image, w_conv1) + b_conv1)
h_pool1 = max_pool_2x2(h_conv1)
# second convolutional layer
w_conv2 = weight_variable([5, 5, 32, 64])
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1, w_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)
# densely connected layer
w_fc1 = weight_variable([7 * 7 * 64, 1024])
b_fc1 = bias_variable([1024])
h_pool2_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 64])
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, w_fc1) + b_fc1)
# dropout
keep_prob = tf.placeholder("float")
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
# readout layer
w_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])
y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop, w_fc2) + b_fc2)
# train and evaluate the model
cross_entropy = -tf.reduce_sum(y_ * tf.log(y_conv))
# train_step = tf.train.AdagradOptimizer(1e-4).minimize(cross_entropy)
train_step = tf.train.GradientDescentOptimizer(1e-3).minimize(cross_entropy)
correct_prediction = tf.equal(tf.argmax(y_conv, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))
sess.run(tf.global_variables_initializer())
for i in range(20000):
batch = mnist.train.next_batch(50)
if i % 100 == 0:
train_accuracy = accuracy.eval(feed_dict={x: batch[0], y_: batch[1], keep_prob: 1.0})
print("step %d, train accuracy %g" % (i, train_accuracy))
train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
# 視訊記憶體不足錯誤
# print("test accuracy %g" % accuracy.eval(feed_dict={x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0}))
a = 10
b = 50
sum = 0
for i in range(a):
testSet = mnist.test.next_batch(b)
c = accuracy.eval(feed_dict={x: testSet[0], y_: testSet[1], keep_prob: 1.0})
sum += c * b
#print("test accuracy %g" % c)
print("test accuracy %g" % (sum / (b * a)))
=================================================
anaconda安裝opencv-python
在Linux下，配置Anaconda的環境變數，請參考我的另一篇文章。
以上工作都完成之後，接下來開始輸入命令了。
conda install --channel https://conda.anaconda.org/menpo opencv3
如果不行試試這個
conda install -c https://conda.anaconda.org/menpo opencv3
樓主親測，第一種方法就行得通。
接著就是等待安裝了。Linux下詢問yes/no的時候，輸入yes。
記憶兩個比較實用的conda的命令，安裝和更新包：
conda install numpy=1.9.3
conda update numpy=1.9.3
========================
conda install -c menpo opencv最後測試一下，沒問題。
import cv2
img = cv2.imread(r'F:\Pictures\tiger.jpg')
cv2.imshow('img', img)
cv2.waitKey(0)
cv2.destroyAllWindows()