pycalceff 1.0.1

Calculate binomial efficiencies and their uncertainties

pycalceff

A Python project for calculating (binomial) efficiencies and their uncertainties. The mathematical theory and derivation of the formulas can be found in Fermilab Technical Memo 2286-cd. If you use this software for published work, please cite this note.

The default algorithm for finding the shortest interval is based on Hyndman, R. J. (1996). Computing and graphing highest density regions, The American Statistician, 50(2), 120-126.

Installation

pip install pycalceff

Usage

# Show help
pycalceff --help

# Show version
pycalceff --version

# Calculate efficiencies from data file with 95% confidence intervals
# Displays results in a formatted table
pycalceff data.txt 0.95

# Save results to a tab-separated values (TSV) file
pycalceff --out results.tsv data.txt 0.95

# Save results to a comma-separated values (CSV) file
pycalceff --out results.csv --use-csv data.txt 0.95

# Error: --use-csv requires --out
pycalceff --use-csv data.txt 0.95  # This will show an error

Output Formats

• Console (default): Results are displayed in a nicely formatted table using Rich.
• TSV File: Tab-separated values with full precision scientific notation. Includes header row with columns: k, n, mode, low, high.
• CSV File: Comma-separated values with full precision scientific notation. Includes header row with columns: k, n, mode, low, high.

Data File Format

The input file should contain lines with two integers each: number of successes (k) and number of trials (n). Lines beginning with '#' are treated as comments and ignored.

Example input file:

# Example efficiency data
# Format: k n (successes, trials)
10 20
5 15
0 10

Example console output:

Efficiency Results                   
┏━━━━┳━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃  k ┃  n ┃         Mode ┃          Low ┃         High ┃
┡━━━━╇━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ 10 │ 20 │ 5.000000e-01 │ 3.332780e-01 │ 6.667220e-01 │
│  5 │ 15 │ 3.333333e-01 │ 1.092969e-01 │ 6.240037e-01 │
│  0 │ 10 │ 0.000000e+00 │ 0.000000e+00 │ 3.085212e-01 │
└────┴────┴──────────────┴──────────────┴──────────────┘

Example TSV file content (results.tsv):

k	n	mode	low	high
10	20	5.00000000000000000e-01	3.33278000000000000e-01	6.66722000000000000e-01
5	15	3.33333333333333315e-01	1.09296900000000000e-01	6.24003700000000000e-01
0	10	0.00000000000000000e+00	0.00000000000000000e+00	3.08521200000000000e-01

Developer Guide

This section describes how to set up a development environment to contribute to pycalceff.

Development Setup

To set up the development environment, you will need to have Miniforge or Miniconda installed.

1. Clone the repository:

   git clone https://github.com/marcpaterno/pycalceff.git
   cd pycalceff
   

2. Create and activate the development environment: The Makefile provides a convenient target for this.

   make setup
   conda activate pycalceff-dev
   

   The setup target uses the environment-dev.yml file to create a consistent Conda environment named pycalceff-dev with all necessary build and development tools.

3. Install the package in editable mode: Once the environment is active, install pycalceff in editable mode with all development dependencies.

   make install
   

   This target uses pip to install the project in "editable" mode (-e), so any changes you make to the source code are immediately reflected when you run the tool. It also installs the [dev] dependencies listed in pyproject.toml.

Development Workflow

The Makefile contains targets for common development tasks.

• Run all checks: To ensure code quality, run all formatting, linting, type-checking, and unit tests at once.

  make check
  

• Run unit tests: To run the test suite with pytest.

  make test
  

• Format code: To automatically format code with ruff.

  make format
  

Packaging and Release Workflow

The build system is configured to produce PyPI packages.

1. Build the packages: To build the sdist and wheel for PyPI, use the build target.

   make build
   

   The PyPI artifacts will be in the dist/ directory.

2. Test local package installation: Before publishing, you can test the PyPI package in a clean, isolated environment without uploading it anywhere. This is the most important verification step.

   make test-install
   

   This command will:

   • Build the packages.
   • Create a temporary environment and install the PyPI wheel, then run tests.
   • Clean up the temporary environment afterward.