checkrs

Tools for simulation-based model checking.

Description

The checkrs package contains functions for creating 7 model checking/diagnostic plots described in

Brathwaite, Timothy. "Check yourself before you wreck yourself: Assessing discrete choice models through predictive simulations" arXiv preprint arXiv:1806.02307 (2018). https://arxiv.org/abs/1806.02307.

Beyond the plots described in this paper, checkrs enables the creation of reliability and marginal model plots that use continuous scatterplot smooths based on Extremely Randomized Trees as kernel estimators, as opposed to only allowing discrete smooths based on binning.

As for the name, checkrs is a play on the word "checkers," i.e., those tools one uses to check, or one who checks. The name is also a play on the phrases "check the research of scientists" and "check research scientists."

Installation

pip install checkrs

Usage

Note that example_project is fictitious! This example is, literally, just an example.

from checkrs import ChartData, ViewSimCDF

from example_project import load_data

design, targets_observed, targets_simulated = load_data()

chart_data = ChartData.from_raw(
  targets=targets_observed,  # 1D Ndarray or Tensor
  targets_simulated=targets_simulated, # 2D Ndarray or Tensor
  design=design # DataFrame or None
)

chart = ViewSimCDF.from_chart_data(chart_data)

chart_plotnine = chart.draw(backend="plotnine")
chart_altair = chart.draw(backend="altair")

####
## Save to a variety of formats
####
# chart.save("temp_plot.png")
# chart.save("temp_plot.pdf")
# chart.save("temp_plot.json")
# chart.save("temp_plot.html")

See docstrings for ChartData.from_raw, ViewSimCDF.from_chart_data, and ViewSimCDF.save.

To-Do:

• Add package to conda and conda-forge

Development installation

To work on and edit checkrs, the following setup process may be useful.

1. from the project root, create an environment checkrs with the help of conda,

   cd checkrs
   conda env create -n checkrs -f environment.yml
   

2. activate the new environment with

   conda activate checkrs
   

3. install checkrs in an editable fashion using:

   flit install --pth-file
   

Optional and needed only once after git clone:

4. install several pre-commit git hooks with:

   pre-commit install
   

   and checkout the configuration under .pre-commit-config.yaml. The -n, --no-verify flag of git commit can be used to deactivate pre-commit hooks temporarily.

Then take a look into the scripts and notebooks folders.

Dependency Management & Reproducibility

1. Always keep your abstract (unpinned) dependencies updated in environment.yml, requirements.in, and eventually in pyproject.toml if you want to ship and install the package via pip later on.

   • Use environment.yml for dependencies that cannot be installed via pip.
   • Use requirements.in for dependencies that can be installed via pip.
   • Use pyproject.toml for dependencies that are needed for checkrs to function at all, not just in development.

2. Create concrete dependencies as requirements.txt for the exact reproduction of your environment with:

   pip-compile requirements.in
   

3. Manually update any non-pip dependencies in environment.yml, being sure to pin any such dependencies to a specific version.

4. Update your current environment using:

   conda env update -f environment.yml
   

   Or

   pip install -r requirements.txt
   

   if you did not update any non-pip dependencies.

Project Organization

├── AUTHORS.rst             <- List of developers and maintainers.
├── CHANGELOG.rst           <- Changelog to keep track of new features and fixes.
├── LICENSE.txt             <- License as chosen on the command-line.
├── README.md               <- The top-level README for developers.
├── configs                 <- Directory for configurations of model & application.
├── data
│   ├── external            <- Data from third party sources.
│   ├── interim             <- Intermediate data that has been transformed.
│   ├── processed           <- The final, canonical data sets for modeling.
│   └── raw                 <- The original, immutable data dump.
├── docs                    <- Directory for Sphinx documentation in rst or md.
├── environment.yaml        <- The conda environment file for reproducibility.
├── models                  <- Trained and serialized models, model predictions,
│                              or model summaries.
├── notebooks               <- Jupyter notebooks. Naming convention is a number (for
│                              ordering), the creator's initials and a description,
│                              e.g. `1.0-fw-initial-data-exploration`.
├── references              <- Data dictionaries, manuals, and all other materials.
├── reports                 <- Generated analysis as HTML, PDF, LaTeX, etc.
│   └── figures             <- Generated plots and figures for reports.
├── scripts                 <- Analysis and production scripts which import the
│                              actual PYTHON_PKG, e.g. train_model.
├── setup.cfg               <- Declarative configuration of your project.
├── setup.py                <- Use `python setup.py develop` to install for development or
|                              or create a distribution with `python setup.py bdist_wheel`.
├── src
│   └── checkrs             <- Actual Python package where the main functionality goes.
├── tests                   <- Unit tests which can be run with `py.test`.
├── .coveragerc             <- Configuration for coverage reports of unit tests.
├── .isort.cfg              <- Configuration for git hook that sorts imports.
└── .pre-commit-config.yaml <- Configuration of pre-commit git hooks.

Note

This project has been set up using PyScaffold 3.3a1 and the dsproject extension 0.4. For details and usage information on PyScaffold see https://pyscaffold.org/.