Week 2: 1 step back, but 2 steps forward

Mar 02, 2019

At the beginning of this week, I ran into an issue where I couldn’t figure out how to convert some of my R code into Python. After speaking to my advisors and researching methods online, we decided to switch back to R because the issue was proving more complicated than we originally thought. Despite this setback, I made some good progress this week!

Given the large amounts of data I am using to train and test my neural networks, I need more computing power than my computer can offer. Therefore, I’m using the University of Iowa’s Argon Cluster to run my code. In order to interact with the Argon Cluster, I’m using Terminal and SecureCRT (a user interface). Since I’m not a student at the University of Iowa, I didn’t have access to the Argon cluster, but after calling them, they were willing to make an exception.

While my account was being processed, I worked on modifying my code so that my neural network could classify all 10 digits (0-9) instead of just 2. Lastly, I read Classification and Regression by randomForest to prepare for future regression styles I may use and studied up basic Linux commands to help me use Terminal effectively.

That’s all for this week!

2 Replies to “Week 2: 1 step back, but 2 steps forward”

  1. Vidur G. says:

    Nice progress! Sometimes we have to completely switch technologies when hitting roadblocks.
    Is there a specific reason why you chose Python over R earlier?

  2. What was your experience working with the Argon Cluster?

Leave a Reply

Your email address will not be published.