vse-sim 0.1.1

Voter Satisfaction Efficiency simulation tools for voting systems.

VSE Sim

VSE Sim provides Python tools for running Voter Satisfaction Efficiency (VSE) simulations for voting methods.

The package is published as vse-sim and imports as vse_sim. The original top-level modules, such as vse and voterModels, are still installed for older scripts and examples.

For background on the metric and published simulation results, see the Voter Satisfaction Efficiency FAQ.

Install

Install the released package from PyPI:

python -m pip install vse-sim

In a notebook, use %pip so the package is installed into the active kernel:

%pip install vse-sim

For reproducible notebooks or environments, pin a released version:

python -m pip install "vse-sim==0.1.1"

To install the latest code from GitHub instead of PyPI:

python -m pip install "vse-sim @ git+https://github.com/wclark/vse-sim.git@main"

Basic Usage

Prefer the modern vse_sim namespace for new code:

import vse_sim as vse
from vse_sim import Mav, PolyaModel, Score, baseRuns, medianRuns
from vse_sim.debug_dump import setDebug

setDebug(False)

results = vse.run_simulation(
    PolyaModel(),
    [[Score(), baseRuns], [Mav(), medianRuns]],
    nvot=5,
    ncand=4,
    niter=3,
    seed="quickstart",
)

results.frame.head()

Work with results using pandas-native helpers:

summary = results.summarize(group_by="method")
leaderboard = results.leaderboard()
report_tables = results.report()
axes = results.plot_vse(group_by="method", kind="barh")

CsvBatch remains available when you want the legacy batch object or metadata CSV writer:

from vse_sim import CsvBatch

csvs = CsvBatch(
    PolyaModel(),
    [[Score(), baseRuns], [Mav(), medianRuns]],
    nvot=5,
    ncand=4,
    niter=3,
    seed="quickstart",
)
csvs.saveFile("quickstart-results")

Legacy imports remain supported:

from vse import CsvBatch
from voterModels import PolyaModel

See Installation and notebook usage for environment setup, including GitHub installs and notebook workflow notes. For a broader copy-paste cookbook that exercises the package surface from Jupyter, see Jupyter notebook examples.

Development

Create or activate a Python 3.10+ environment, then install the project in editable mode with development tools:

python -m pip install -e ".[dev,publish]"

Install local Git hooks if you want pre-commit checks:

pre-commit install

Run the default local quality gate:

nox

The default Nox gate validates metadata, runs Ruff format/lint checks, runs the Python 3.10 test and coverage suite, builds the package, validates distribution metadata and wheel contents, and audits dependencies.

Run Python 3.12 tests explicitly when that interpreter is available:

nox -s tests-3.12

Useful direct commands:

python -m pytest --doctest-modules --cov=. --cov-fail-under=100
python -m ruff format --check .
python -m ruff check .
python -m build
python -m twine check dist/*
check-wheel-contents dist/*.whl
python -m pip_audit --skip-editable --progress-spinner off .

Coverage reports are written to htmlcov/, coverage.xml, and pytest-results.xml when the full coverage command is run.

Repository Layout

• vse_sim/: modern package facade for new imports.
• Root Python modules: legacy-compatible modules that remain importable.
• test/: coverage and compatibility tests.
• data/: retained legacy/reference data artifacts.
• docs/: GitHub Pages content plus install and release notes.

See Release process for the PyPI publishing workflow.