Prerequisites

• Docker Compose

We're using docker for local dev as python development on windows is poopy doo doo

docker-compose up -d

Shell into python docker (can also use shell.ps1)

docker exec -it stonks_nlp /bin/bash

Generating Model

trainer.py is a script to generate an sklearn model under stonks_model (constants.model_filename) todo: We'll start committing this file

python src/trainer.py

Stonks cli

CLI for getting sentiment of tweet

python -m src.stonks_cli "some real tweet"

Will print out an integer sentiment

Testing

We're using python's unittest for tests. Shell in (you can use shell.ps1) and run:

python -m unittest {testfile}