fgmutils 1.0.0

Forest Growth Model Utilities

FGMUtils (Python Port)

Version 0.9.6

Overview

This folder contains the Python implementation of the FGMUtils toolkit. It mirrors the behaviour of the original R package while exposing a modern packaging layout (src/, pyproject.toml) suitable for publication on PyPI. The modules cover data preparation, prediction, evaluation, statistics, and advanced volume analysis, all validated against fixtures generated by the R tests.

Objectives

• Provide a faithful port of the R algorithms used for forest growth and production modelling.
• Deliver reusable building blocks (dataops, prediction, evaluation, metrics, volume) that can be imported in analytical pipelines.
• Ensure numerical parity with the R package through pytest suites and R-driven fixtures under tests/data.
• Align metadata (name, version, authors, license) with the source R package so that releases stay synchronized.

Key Modules & Functions

• dataops: utilities such as separa_dados, cria_dados_pareados, and atualiza_campo_base.
• prediction: resilient predizer implementation compatible with statsmodels fits.
• evaluation: avalia_estimativas and eval_age_based, plus ranking helpers.
• metrics: MAE, RMSE, corrected RRMSE, CE, bias, syx, mspr, r21a, r29a.
• volume: avalia_volume_avancado, avalia_volume_age_based, and calcula_volume_default.
• project: mirrors projectBaseOriented to generate projections by age class.

Authors

The Python distribution inherits the authorship from the R package:

• Clayton Vieira Fraga Filho (maintainer)
• Ana Paula Simiqueli
• Gilson Fernandes da Silva
• Miqueias Fernandes
• Wagner Amorim da Silva Altoe

Testing & Publication

• Execute pytest from fgmutils_python/ to run the parity test suite.
• Metadata is synced from fgmutils_R/DESCRIPTION via scripts/sync_metadata.py; run this script whenever the R package information changes before building wheels (python -m build).