There is a newer version of the record available.

Published September 18, 2025 | Version v1.0.6
Software Restricted

manwestc/TINTOlib-Python-Library: TINTOlib v1.0.6.1 – Categorized Methods, Unified Training API, and More

Authors/Creators

  • 1. @oeg-upm

Description

✅ TINTOlib v1.0.6 – Categorized Methods, Unified Training API, and More

🚀 What's new in TINTOlib v1.0.6

✨ Methods Categorized
All methods are now grouped into two categories:

  • Parametric methods: TINTO, IGTD, and REFINED
  • Non-parametric methods: BarGraph, DistanceMatrix, Combination, SuperTML, FeatureWrap, and BIE

✨ Unified Training API
The legacy generateImages() function has been removed.
We now follow a standardized approach:

  • Parametric methods use: .fit() and .transform()
  • Non-parametric methods use: .fit_transform()

This improves consistency and reduces the risk of data leakage in real-world pipelines.

✨ New Parameter for REFINED
A new normalize parameter has been added to the REFINED method, improving both performance and flexibility.

✨ Updated Notebooks
All example Jupyter Notebooks in the TINTOlib Crash Course have been updated to reflect these changes, including examples for both TensorFlow and PyTorch.

Files

Restricted

The record is publicly accessible, but files are restricted. <a href="https://zenodo.org/account/settings/login?next=https://zenodo.org/records/17153147">Log in</a> to check if you have access.

Additional details

Related works

Is supplement to
Software: https://github.com/manwestc/TINTOlib-Python-Library/tree/v1.0.6 (URL)

Software

Repository URL
https://github.com/manwestc/TINTOlib-Python-Library