Improved interaction with exposed models #114

	name: Pull Request

	on:
	workflow_dispatch:
	pull_request:
	branches:
	- main

	jobs:
	test:
	runs-on: ubuntu-latest
	services:
	postgres:
	image: postgres:14.3-alpine
	env:
	POSTGRES_DB: postgres
	POSTGRES_PORT: 5432
	POSTGRES_USER: postgres
	POSTGRES_PASSWORD: postgres
	ports:
	- '127.0.0.1:5432:5432'

	steps:
	- uses: actions/checkout@v3
	- uses: actions/setup-python@v4
	with:
	python-version: "3.10"

	- uses: actions/cache@v3
	id: cache-venv
	with:
	path: \|
	~/.local
	.venv
	key: ${{ hashFiles('**/poetry.lock') }}-${{ steps.setup-python.outputs.python-version }}-1

	- name: Install Poetry
	uses: snok/install-poetry@v1
	with:
	virtualenvs-create: false
	version: 1.5.0
	- run: \|
	python -m venv .venv --upgrade-deps
	source .venv/bin/activate
	poetry install --no-interaction --all-extras
	if: steps.cache-venv.outputs.cache-hit != 'true'

	- name: Start MLflow service
	run: \|
	docker network create mlflow \|\| true
	docker run -d --rm \
	--name mlflow \
	--network mlflow \
	-p 8000:8000 \
	ghcr.io/mlflow/mlflow:v2.15.0 \
	/bin/sh -c "mlflow server --backend-store-uri file:///app/mlflow-server/experiments --artifacts-destination file:///app/mlflow-server/artifacts --host 0.0.0.0 --port 8000"

	- name: Verify MLflow server is running
	run: \|
	for i in {1..10}; do
	curl -s http://localhost:8000 && break
	echo "Waiting for MLflow server..."
	sleep 5
	done
	curl http://localhost:8000

	- name: Run tests
	env:
	PSQL_DATABASE_TEST: postgresql://postgres:postgres@localhost:5432/postgres
	MLFLOW_TRACKING_URI: http://localhost:8000
	MLFLOW_REGISTRY_URI: http://localhost:8000
	run: \|
	source .venv/bin/activate
	pytest

Provide feedback