Developed by Christopher Blanton, Ph.D.

Georgia Institue of Technology

Version 9/29/2020

You have learned some theory but how do you do your first Machine Learning project can be challenging:

• How to handle the data?
• How to identify what is essential in the data?
• How to visualize correlations?
• How to evaluate your model?
• What are some common pitfalls and how can you avoid falling into them?

In this workshop, we will work through example first project in machine learning. We will begin with a concept of a project, ingest the data, visualize and view potential correlations, select, train our model, and evaluate the model.

Prerequisites: Some familiarity with Python.

Here are the quick steps to starting the workshop

1. Connect to the cluster using

   $ ssh userid@pace-ice.pace.gatech.edu
   

2. Get the materials for the workshop using

   $ git clone https://github.com/chrisblanton/gatech_machine_learning.git
   

3. Change to the directory using

   $ cd ./gatech_machine_learning
   

4. Start the session by running the quickstart script

   $ chmod +x ./start_workshop.sh
   $ ./start_workshop.sh
   

5. Follow the commands on-screeen as seen below.

The course materials are available at

https://github.com/chrisblanton/gatech_machine_learning

The materials can be obtained by cloing the git repo

$ git clone https://github.com/chrisblanton/gatech_machine_learning.git

or by using the button to download a ZIP file.

The job can be started using the wrapper script as

$ pace-jupyter-notebook -l nodes=1:ppn=1 -l walltime=02:00:00 -q pace-training  \
  --anaconda=anaconda3/2019.10 --conda-env=/storage/home/hpaceice1/shared-classes/envs/mlws

Then following the instructions one the job has begun:

     Job successfully submitted!
     Waiting for job to start...
     
     Starting jupyter notebook...

     Connect to your jupyter notebook via the following steps:
     1) Press SHIFT + ~ then SHIFT + C to open an SSH console (The prompt 'ssh>' should appear on the next line)
     	 ***Note: '~' MUST be the first character on the line to   	  be recognized as the escape character, in which case it          will not appear on your terminal.***
      	 ***If you see the '~' character when you start typing, delete it, hit 'ENTER' and type 'SHIFT' + '~' + 'C' again.***
   	 2) Type -L 55084:rich133-c32-10-r:55084 and then ENTER
   	 3) Connect your browser to http://localhost:55084/?token=ccd58369799be30f3173967b4aed5fb2a0029e5381a1af90

After copying and pasting the link, you should reach a webpage with a directory listing. You should be able to navigate to the directory with the course materials

Please complete the quick survey.

https://gatech.co1.qualtrics.com/jfe/form/SV_55uzMYLufTuiLch