I spoke to Professor Cho to get guidance on my next steps. I told him about my current progress. He told me to start working on my write up. I don't remember verbatim what he said, but essentially I need to talk about what I did, some challenges, and next steps. One of the specifics he mentioned was to provide numbers on performance. For example, out of the total number of input data points, how many do I have parsed? (currently I only do points that are "no parking" and have a start and end time) Of those parsed, how many are accurate? Random sampling is the name of the game here.
At the very least, I want to set up a system where I've NLP parsed the entire dataset. Since I can't get it to work all at once, I'm going to break it up into chunk sizes I know works, 500. I'll need to build some additional pipe-age for this.
No comments:
Post a Comment