Writing your First Testing Suite with pytest

Overview

Repository contains sample testing suite to illustrate how to write your first testing suite and how to do so with pytest.

We've all heard that we should be testing our Machine Learning models in production and getting good test coverage. This repository and referenced talk will give an overview of 4 types of tests: unit, integration, regression and parametrized test, via examples in pytest, to help you write more robust code in no-time, no prior testing experience necessary.

Associated Slide Decks

• IDEAS webinar (2019-07-20) slide deck
• PyLadies Los Angeles meetup (2018-11-08) slide deck

Prerequisites to Running the Code/Tests

1. Prerequisites: Install Python 3.6+

2. Clone (or download) this repository

3. (Optionally) Start a virtual environment for the project

4. Install required package pytest (to run tests) and pytest-cov (to get code coverage)

pip3 install -r requirements.txt 

Run the Code

From root of the folder, run:

python3 main.py 100 .10 -.2

Run the Tests

From root of the folder, run:

pytest tests/. -x --pdb

Check code coverage

From root of the folder, run:

pytest tests/. --cov --cov-report=html
open htmlcov/index.html