Add `skip_noisy_assignment` to `dataset.cluster` #2381

	name: Backend
	on:
	pull_request:
	types: [opened, reopened, synchronize]
	paths:
	- 'lilac/**'
	- '*/.py'
	- 'poetry.lock'
	- 'poetry.toml'
	- 'pyproject.toml'
	- 'mypy.ini'
	- 'pytest.ini'
	- '.python-version'
	- '.env'
	- '.coveragerc'
	- '.github/workflows/python.yml'
	- 'scripts/**'

	jobs:
	build:
	name: 'Python lint and test'
	runs-on: ubuntu-latest
	strategy:
	matrix:
	python-version: ['3.9', '3.11']
	timeout-minutes: 15

	steps:
	- uses: actions/checkout@v3
	- name: Install poetry
	run: pipx install poetry
	- name: Set up Python ${{ matrix.python-version }}
	uses: actions/setup-python@v4
	with:
	python-version: ${{ matrix.python-version }}
	cache: 'poetry'
	- name: Install dependencies
	# Re-install pytorch for cpu to avoid errors that cuda was not found
	run: \|
	./scripts/setup_py.sh
	poetry run pip3 install --force-reinstall --pre torch==2.* --index-url https://download.pytorch.org/whl/nightly/cpu
	poetry run pip3 install typing_extensions==4.9.0 --force-reinstall

	- name: Test CLI
	run: \|
	./integration_tests/run_integration_tests.sh

	# We run testing before linting since testing, surprisingly, is much faster.
	- name: Test python
	run: \|
	poetry run ./scripts/test_py.sh

	- name: Lint python
	run: \|
	poetry run ./scripts/lint_py.sh

Provide feedback