🚀 Project Improvements - Version Bump to 2.3.0

[rafaelpadilla]

⚠️ Review Order Notice

Although this PR includes the changes introduced in PR #180 (proposing v2.2.0), it is recommended that the current PR be reviewed after PR #180.
This ordering is intentional and aligns with the planned sequence of improvements.

Overview

This pull request delivers a set of refactors and improvements aimed at modernizing path handling, clarifying feature-extractor behavior, and introducing structured logging for better observability and experiment traceability. The scope of this delivery focuses on consistency, maintainability, and operational visibility.

Given the scope and impact of these improvements, this pull request suggests a bump to v2.3.0 (not v2.2.0).

Summary of Changes

1. Use pathlib.Path in place of os.path for modern path management

All occurrences of os.path were replaced with pathlib.Path across the codebase to provide a modern, object-oriented interface for path handling. This improves consistency, readability, and alignment with current Python best practices.

2. Feature extractors: is_windowed as instance attribute

The unnecessary use of getattr(config, "is_windowed", False) was removed from the feature extraction classes (ExtractStatisticalFeatures, ExtractEWStatisticalFeatures, ExtractWaveletFeatures). The default is now self.is_windowed = False on the instance, and error messages were updated to direct users to set extractor.is_windowed = True on the instance when needed. Tests were adjusted to set is_windowed on extractor instances rather than via config, making it explicit that is_windowed is an instance attribute, not a config parameter.

3. Structured logging system

A structured logging system was introduced across the framework to improve observability, debugging, and experiment traceability. A logging directory was defined in constants.py, a logging_config.py setup file was added to configure format, handlers, and automatic log file creation, and the logger setup runs when the toolkit is imported. Loggers were added to trainer/trainer.py and models/mlp.py. Each run creates a log file under the configured directory following the pattern run_<timestamp>.log, with entries in the format <timestamp> | <level> | <module> | <message>.

4. Include youtube video links to README.md files

YouTube links were added to README.md files for 3W dataset resources used during workshop presentation.

---

Final Notes

These changes focus on modern path handling, clearer contracts for feature extractors, and better visibility into toolkit execution through logging, strengthening both code consistency and the ability to debug and trace experiments.

Thank you for reviewing this delivery.
🎉🥳🎊🚀✨

---

By creating this pull request, I confirm that I have read and fully accept and agree with one of the Petrobras' Contributor License Agreements (CLAs):

• ICLA: Individual Contributor License Agreement on behalf of only yourself;
• CCLA: Corporate Contributor License Agreement on behalf of your employer.

Our CLAs are based on the Apache Software Foundation's CLAs:

• ICLA: Individual Contributor License Agreement
• CCLA: Corporate Contributor License Agreement

[ricardoevvargas]

Hi, @rafaelpadilla.

I hope this PR will result in another significant contribution from the Federal University of Rio de Janeiro to the 3W Project.

We will evaluate it ASAP and will let you know here if we have any questions and/or requests for adjustments.

On behalf of the 3W Community, I thank you for this PR.

PS: the evaluation of PR #180 is still ongoing.

[ricardoevvargas]

Hi, @rafaelpadilla.

Can you please check what needs to be done to resolve the identified conflicts?

[rafaelpadilla]

Hi @ricardoevvargas ,

✔️ No conflicts now.

[ricardoevvargas]

Hi, @rafaelpadilla.

Great. Once again and on behalf of the 3W Community, I thank you for this PR.