This article is about how to build TensorFlow under Aliyun Student Server

OS Ubuntu 16.04 64-bit python 3.5

1. Updating Software Sources

$ apt-get update

2. Install Python 3.5

$ apt-get install python3.5
$ cp /usr/bin/python /usr/bin/python_bak #backups
$ rm /usr/bin/python #delete
$ ln -s /usr/bin/python3.5 /usr/bin/python #Set default to Python 3.5

Enter the python command to view the current default Python version

3. Install TensorFlow

$ apt-get install python3-pip 
$ pip3 install tensorflow  # Python 3.n; CPU support (no GPU support)

locale.Error: unsupported locale setting problem (language environment configuration problem) may be encountered when installing TensorFlow with pip3

Solution

Step 1

$ locale
locale: Cannot set LC_ALL to default locale: No such file or directory
LANG=en_US.UTF-8
LANGUAGE=
LC_CTYPE="en_US.UTF-8"
LC_NUMERIC=zh_CN.UTF-8
LC_TIME=zh_CN.UTF-8
LC_COLLATE="en_US.UTF-8"
LC_MONETARY=zh_CN.UTF-8
LC_MESSAGES="en_US.UTF-8"
LC_PAPER=zh_CN.UTF-8
LC_NAME=zh_CN.UTF-8
LC_ADDRESS=zh_CN.UTF-8
LC_TELEPHONE=zh_CN.UTF-8
LC_MEASUREMENT=zh_CN.UTF-8
LC_IDENTIFICATION=zh_CN.UTF-8
LC_ALL=

Step 2

$ export LC_ALL=C
root@ubuntu:~# locale
LANG=en_US.UTF-8
LANGUAGE=
LC_CTYPE="C"
LC_NUMERIC="C"
LC_TIME="C"
LC_COLLATE="C"
LC_MONETARY="C"
LC_MESSAGES="C"
LC_PAPER="C"
LC_NAME="C"
LC_ADDRESS="C"
LC_TELEPHONE="C"
LC_MEASUREMENT="C"
LC_IDENTIFICATION="C"
LC_ALL=C

Step 4 Verifies TensorFlow Installation

$ python #Enter python
import tensorflow as tf
hello = tf.constant('Hello TensorFlow!')
sess = tf.Session()
print(sess.run(hello))
exit() #Sign out

Using Tensorboard Visualization Tool

1. Install graphical interface for Aliyun server

$ apt-get install x-window-system-core
$ apt-get install gnome-core
$ apt-get install gdm
$ startx  #Aliyun console remote connection can see the graphical interface

Install vim to edit documents

$ apt-get install vim

If E: Sub-process/usr/bin/dpkg returned an error code appears when apt-get is installed

Solution

sudo mv /var/lib/dpkg/info /var/lib/dpkg/info.bak //Now rename the info folder
sudo mkdir /var/lib/dpkg/info //Create a new info folder
sudo apt-get update

2. Create a new TensorFlow folder under the home directory and put it into the tensorboard.py file.

$ mkdir TensorFlow
$ cd TensorFlow
$ vim tensorboard.py

tensorboard.py file

"""
Please note, this code is only for python 3+. If you are using python 2+, please modify the code accordingly.
"""
from __future__ import print_function
import tensorflow as tf
import numpy as np


def add_layer(inputs, in_size, out_size, n_layer, activation_function=None):
    # add one more layer and return the output of this layer
    layer_name = 'layer%s' % n_layer
    with tf.name_scope(layer_name):
        with tf.name_scope('weights'):
            Weights = tf.Variable(tf.random_normal([in_size, out_size]), name='W')
            tf.summary.histogram(layer_name + '/weights', Weights)
        with tf.name_scope('biases'):
            biases = tf.Variable(tf.zeros([1, out_size]) + 0.1, name='b')
            tf.summary.histogram(layer_name + '/biases', biases)
        with tf.name_scope('Wx_plus_b'):
            Wx_plus_b = tf.add(tf.matmul(inputs, Weights), biases)
        if activation_function is None:
            outputs = Wx_plus_b
        else:
            outputs = activation_function(Wx_plus_b, )
        tf.summary.histogram(layer_name + '/outputs', outputs)
    return outputs


# Make up some real data
x_data = np.linspace(-1, 1, 300)[:, np.newaxis]
noise = np.random.normal(0, 0.05, x_data.shape)
y_data = np.square(x_data) - 0.5 + noise

# define placeholder for inputs to network
with tf.name_scope('inputs'):
    xs = tf.placeholder(tf.float32, [None, 1], name='x_input')
    ys = tf.placeholder(tf.float32, [None, 1], name='y_input')

# add hidden layer
l1 = add_layer(xs, 1, 10, n_layer=1, activation_function=tf.nn.relu)
# add output layer
prediction = add_layer(l1, 10, 1, n_layer=2, activation_function=None)

# the error between prediciton and real data
with tf.name_scope('loss'):
    loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),
                                        reduction_indices=[1]))
    tf.summary.scalar('loss', loss)

with tf.name_scope('train'):
    train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

sess = tf.Session()
merged = tf.summary.merge_all()

writer = tf.summary.FileWriter("logs/", sess.graph)

init = tf.global_variables_initializer()
sess.run(init)

for i in range(1000):
    sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
    if i % 50 == 0:
        result = sess.run(merged,
                          feed_dict={xs: x_data, ys: y_data})
        writer.add_summary(result, i)

# direct to the local dir and run this in terminal:
# $ tensorboard --logdir logs

Run command

$ python tensorboard.py #Generate a logs folder and enter the tensorboard command to view the tensorboard web site
$ tensorboard --logdir logs

Enter 127.0.0.1:6006 in the browser to visit Tensorboard Visual Page of tensorflow.py

Tensorboard error TensorBoard attempted to bind to port 6006, but it was already in use solution

$ lsof -i:6006
root@iZ2ze2v60tavfuwqu1ipvzZ:~/TensorFlow# lsof -i:6006
COMMAND    PID USER   FD   TYPE DEVICE SIZE/OFF NODE NAME
tensorboa 1635 root    3u  IPv4  19343      0t0  TCP *:x11-6 (LISTEN)

Kill process

root@iZ2ze2v60tavfuwqu1ipvzZ:~/TensorFlow# kill -9 1635
[1]+  Killed                  tensorboard --logdir logs

Run again

$ tensorboard --logdir logs

jupyter notebook installation configuration

$ pip install jupyter

$ jupyter notebook –generate-config –allow-root
$ ipython

Python 3.5.2 (default, Aug  4 2017, 02:13:48) 
Type 'copyright', 'credits' or 'license' for more information
IPython 6.1.0 -- An enhanced Interactive Python. Type '?' for help.

In [1]: from notebook.auth import passwd
In [2]: passwd()
Enter password: 
Verify password: 
Out[2]: 'token' //token refers to a string of generated text that needs to be used later
In [3]: exit()

Modify the jupyter notebook configuration file

$ vim ~/.jupyter/jupyter_notebook_config.py

c.NotebookApp.ip='*'
c.NotebookApp.password = u'token'
c.NotebookApp.open_browser = False
c.NotebookApp.port =8888    #Specify a port at will, or use the default 8888

Visit jupyter notebook

$ jupyter notebook  --ip=0.0.0.0 --no-browser --allow-root

Browser input 0.0.0.0:8888 access

If you think Aliyun backstage remote connection server is too slow, you can choose to download and install teamviewer 12 in the graphical interface for remote desktop control.

End