As the README states, I'm not exactly sure how far I will take this analysis. I hit a wall in creating an oil well detector and so want to see how Maya prepared here data for detecting ships. She is using an existing and already labeled dataset of ships from kaggle - https://www.kaggle.com/rhammell/ships-in-satellite-imagery/downloads/ships-in-satellite-imagery.zip/9
Even will non-CNN based algorithms, she has reasonably high success with her ship detector, and so I wanted to understand how the ship dataset differs from my more limited dataset and how her data preparation has differed from my own.
Here goes nothing.