Changelog

For the complete version history, see the main CHANGELOG.md on GitHub.

Latest Release: v0.0.4 (2026-02-12)

Major Features

Preprocessing Pipeline

  • Complete FTIRdataprocessing class with evaluation-first approach

  • 50+ baseline correction algorithms via pybaselines

  • 7 denoising methods with comprehensive evaluation

  • 17+ normalization methods with classification-based selection

  • Atmospheric correction for CO₂ and H₂O interference

Machine Learning & Analysis

  • Complete FTIRdataanalysis class for classification and analysis

  • 20+ machine learning models (XGBoost, LightGBM, CatBoost, SVM, etc.)

  • Model explainability with SHAP values

  • Hyperparameter tuning for top models

  • Dimensionality reduction: PCA, t-SNE, UMAP, PLS-DA, OPLS-DA

Bundled Datasets

  • 6 FTIR plastic datasets from published studies (2018-2024)

  • Jung et al. 2018 (~500 spectra)

  • Kedzierski et al. 2019 (2 variants, ~4,000 spectra)

  • Frond et al. 2021 (~400 spectra)

  • Villegas-Camacho et al. 2024 (C4 and C8, ~6,000 spectra)

Documentation & Examples

  • Comprehensive user guide with step-by-step tutorials

  • Interactive Jupyter notebooks for method selection

  • Complete API reference

  • Real-world examples and use cases

Improvements

  • Enhanced visualization capabilities across all modules

  • Better error handling and validation

  • Support for both Pandas and Polars DataFrames

  • Improved memory management for large datasets

  • Professional packaging for PyPI distribution

For older versions and detailed changes, see CHANGELOG.md.