pimlottc-gov/demand-modeler

Demand Modeller

https://fearless-dreamer.herokuapp.com/

Table of Contents

• Setup
• Bootstrapping Heroku
• Loading new data
  • csbs
  • USA Facts

Setup

1. Download the TIGER files:

wget https://www2.census.gov/geo/tiger/TIGER2019/STATE/tl_2019_us_state.zip -O data/tl_2019_us_state.zip
wget https://www2.census.gov/geo/tiger/TIGER2019/COUNTY/tl_2019_us_county.zip -O data/tl_2019_us_county.zip

1. Unzip the files (there should be a .shp, .shx, .dbf, .prj file for each)
2. install Docker

docker-compose up postgres

1. load tiger files into the database

docker build -f loader.Dockerfile -t loader . && docker run -it --network=demand-modeler_default loader

* if the network is not correct double check it with docker network ls

1. start the API

docker-compose up --build api

Bootstrapping Heroku

1. Create a Postgres service and link it to the application

heroku addons:create heroku-postgresql:hobby-dev

1. Load the initial state and county shapefiles into the database using the DATABASE_URL parameter from Heroku

heroku config
export DATABASE_URL="{DBURL from heroku}
make load

1. Install the Go and Nodejs buildpacks

heroku buildpacks:add heroku/nodejs
heroku buildpacks:add heroku/go

1. Deploy

Loading new data

csbs

1. Download the latest CSV file
2. Copy into the data/ directory with this name pattern: `covid19_county_2020_{month}_{day}-{time?}.csv
3. Perform manual cleaning

• Delete New Death column
• Rename 'Walton, FL' to 'Walton County'

1. Run the go application and see if it barfs on the new data. Rinse, repeat until it works.

2. Download USAFacts death dataset

3. Verify top 10 counties in top 10 states in both datasets (USAFacts is source of truth)

4. Verify top-10 state value still matches what the dataset reports.

USA Facts

docker build -f usafacts_loader.Dockerfile -t usafacts . && docker run -it --network=demand-modeler_default usafacts