Transformation

Transformation tools to normalise and stabilise the variance of non-normal data.

Table of Contents

• Overview

  • Installation
  • Demonstration
  • Contributing
  • License
  • Contact

  ---

Overview

normtransform is a high-performance Python package (with Cython/C++ backend) for fitting univariate transformed normal distributions using maximum likelihood or maxium a posteriori (MAP) estimation, with support for left and right censoring. Typically, norm-transform is used to transform variables/data so that they can be used in modelling that assumed normality. The ability to handle scenarios where data may be treated as partially observed (e.g. rainfall) allows significant flexibility in applications.

Interactive Demonstration of normtransform

You can explore an interactive demonstration of the normtransform tool here. This link opens a Jupyter Notebook hosted on the MyBinder platform. Once it loads, click Run-->Run All Cells to execute the code and see the demonstration in action.

Installation

A Python virtual environment is a self-contained directory that includes a specific Python version and any additional packages you install. It helps you isolate dependencies for different projects, ensuring that changes in one environment don’t affect others. This makes it easier to manage project-specific requirements and maintain a clean development setup.

Use the package manager pip to install normtransform.

python -m venv venv # Create a virtual environment
source venv/bin/activate  # Activate the virtual enviroment, on Windows use `venv\Scripts\activate`
pip install normtransform #install normtransform

Once installed users can start with mimicking the demonstration found above within their own environment.

Contributing

Requests are welcome. Please contact the authors below.

License

MIT

Contact

• Andrew Schepen - andrew.schepen@csiro.au
• David Robertson - david.robertson@csiro.au