setup_log.txt

Conda env:
conda create -n traj_scoreboard jupyter pandas django nodejs numpy sqlalchemy psycopg2

or 

conda install jupyter pandas django nodejs numpy sqlalchemy psycopg2

pip install bash_kernel
python -m bash_kernel.install

python BuildDatabase.py  (missing init)

python manage.py migrate
python manage.py inspectdb > benchmark/models.py
replace in models max_length=-1 with max_length=1000
python manage.py createsuperuser


Installing scoreboard from scratch:

Prereqs
Node/NPM
Docker
 Python3 (for database script)
Initial Setup
Get Github OAuth key
In Github
Settings -> Developer settings -> OAuth Apps
Click New OAuth App
Name -> Scoreboard, Homepage -> http://localhost:9001, Authorization callback URL -> http://localhost:9001/auth/github/callback
Click Register Application
Setup Postgres docker container
docker pull postgres
sudo docker run -p 5433:5432 --name scoreboard -e POSTGRES_PASSWORD=scoreboard -e POSTGRES_DB=scoreboard -d postgres (You can change the password)
Create a challenge scoring container
Example at: scoreboard/docker at master · chanzuckerberg/scoreboard · GitHub
Container takes a results text file and prints out json of format
{“data": [0.9, 0.9, 0.9], "additionalData": [[0.95, 0.95, 0.9500000000000001], [0.9, 0.9, 0.9], [0.9, 0.9, 0.9], [0.75, 0.75, 0.75], [0.95, 0.95, 0.9500000000000001], [0.95, 0.95, 0.9500000000000001], [0.95, 0.95, 0.9500000000000001], [0.9, 0.9, 0.9], [0.85, 0.85, 0.85]], "error": "this is an error"}
Build and tag container
docker build -t <containtername> -f Dockerfile .
Setup Config
Database config file see initialize.json
App config file. See scoreboard.cfg.js

Setup environment
(Put this in your bash file for easy access)
export SCOREBOARD_PG_PASSWORD=“postgres”
export SCOREBOARD_PG_USERNAME=“postgres”
export SCOREBOARD_PG_PORT=5433
export SCOREBOARD_GH_APPID=“<gh app id>”
export SCOREBOARD_GH_SECRET=“<gh app secret>"
Build and launch scoreboard
Clone scoreboard git clone https://github.com/chanzuckerberg/scoreboard.git
Checkout production branch git checkout production
Install dependencies npm install
Build client npm run build
Launch server node server
Go to site http://127.0.0.1:9000/