Machine Problem #1 Solution

Description

 

Note: The assignment will be autograded.  It is important that  you do not use additional libraries, or change the provided functions input  and output.

 

Part 1:  Setup

 

• Remote connect to a EWS machine.

 

ssh (netid)@remlnx.ews.illinois.edu

 

 

• Load python module, this will also load pip and virtualenv

 

module load python/3.4.3

 

 

• Create a virtualenv “cs446sp 2018”.

 

virtualenv –system-site-packages ~/cs446sp_2018

 

 

• Activate the virtualenv

 

source ~/cs446sp_2018/bin/activate

 

 

• Update pip

 

pip install –upgrade pip

 

 

• Checkout and change directory to the course svn repository

 

svn co https://subversion.ews.illinois.edu/svn/sp18-cs446/(netid)

cd (netid)

 

 

• Copy mp1 into your svn directory, and change directory to mp1.

 

svn cp https://subversion.ews.illinois.edu/svn/sp18-cs446/_shared/mp1 . cd mp1

 

 

• Install the requirements through  pip.

 

pip install -r requirements.txt

 

 

 

 

 

1

 

2

 

 

Part 2:  Exercise

The purpose of this exercise is to introduce you to basic components of Tensorflow, a Python

machine learning framework which will be used for the assignments  this semester.  Specifi- cally, you will need to implement code that  creates a few basic computation graphs and then executes them.  Before beginning, read through  Tensorflow’s “Getting  Started” tutorial  1

 

The following methods  must  be implemented  (more specific details  can be found in the corresponding  source files):

 

• In run computation.py , the method run computation, which takes  a computation graph node as input,  initializes all variables, runs the node, and returns  the result.

 

• In toy functions.py , the methods toy fn 1, toy fn 2, and toy fn 3, which all build computation graphs for different functions.

 

 

Part 3:  Writing Tests

In  test.py we have provided  basic test-cases.   Feel free to write more.  To test  the code,

run the following (which runs  test.py ):

 

nose2

 

 

 

Part 4:  Submit

Submitting the code is equivalent to committing  the code. This can be done with the follow

command:

 

svn commit -m “Some meaningful comment here.”

 

 

Lastly, double check on your browser that  you can see your code at

 

https://subversion.ews.illinois.edu/svn/sp18-cs446/(netid)/mp1/

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1  https://www.tensorflow.org/get_started/get_started