Our goal is to create a pipeline for predicting MLB games, placing ficticious automatic bets on those games, creating a database to store all of this data, and a dashboard to display our results.
Project Team:
- Rocky Chen
- Kasey Jones
This project relies on a docker container. Locally, this container makes sure all team members are using the same packages. At the server level, it ensures the python scripts run as expected.
From the root directory of the project:
docker-compose build
docker-compose run --rm mlb2019 bash
Sign-up for Amazon web serives and create a postgresql instance. I used the free tier. Amazon RDS Free Tier
Download a way to communicate with the database: pg admin 4, although we will mostly use python to communicate.
Note that publicly available must be yes. Add security group to be custom TCP rule and select anywhere.
hostname: mlb2019.cvsp9ebtej6h.us-east-2.rds.amazonaws.com
port: 5432
username: kaseyriver11
Install docker & and start the daemon:
sudo apt install docker.io
sudo apt install docker-compose
sudo systemctl start docker
sudo docker-compose build
Test Script:
sudo docker-compose run mlb2019 bash -c "python3.6 src/test_script.py"
Setup the cron jobs:
cd mlb2019/
cat docker/cronjobs | crontab -
Make sure there is a local/db_preds file with correct username and password.