Inspiration
We tried to take a bit of a FIFA like ranking system and figure out what "Wins Games." We decided that it breaks down to "Gold" and "Objectives." So we tried to build our ranking system around those.
What it does
First, you download as much data as you can, then filter down to a more manageable size of data for processing. Then build up a ranking value from that data. The eventual output will rate and rank all the available teams according to our algorithm.
How we built it
Python for pulling down the data from AWS. Docker Container to hold all of the code and have easy install. Hadoop for the database. Nuxt 3 for the website.
Challenges we ran into
- Big data is a thing. It took a few weeks to get queries that we could actually use. Also quickly ran out of space on my SSD.
- 1 developer, a few Data Scientists, and a Product Owner. Could have used more devs.
- I was unable to connect the front end to the back end. So many manual steps that I would have liked to automate.
Accomplishments that we're proud of
- WE GOT A PRODUCT OUT THE DOOR!!!
- Docker things, Had a good enough process documented where non-devs were able to get setup with minimal support.
What we learned
Sometimes I gotta take more breaks. A few times after banging my head against a problem, staying up late nights, silently screaming at my computer screen why something was broken. After a good nights rest, solved the problem in under an hour.
What's next for Duffy4Commish
How about a job offer for Duffy as the LCS Commissioner!!! 😉 HAHA, Actually I would like to process data coming from lolesports.com's Live Game Stats. In the past I have set up Postman Calls and could grab that data that is published for the games. Maybe actually finish that, it is in a similar vein.
Log in or sign up for Devpost to join the conversation.