Skip to content

feature: Add ForwardLookingFilter to label_tuning utils#14

Merged
agpenas merged 10 commits into
mainfrom
feat/tuner/forward_looking_filter
Apr 14, 2026
Merged

feature: Add ForwardLookingFilter to label_tuning utils#14
agpenas merged 10 commits into
mainfrom
feat/tuner/forward_looking_filter

Conversation

@agpenas

@agpenas agpenas commented Apr 13, 2026

Copy link
Copy Markdown
Owner

Added

  • Add ForwardLookingFilter postprocessor to filter tuned labels based on forward-looking efficiency.
  • Update README, docs and the label_tuning notebook.

Changed

  • Refactor label tuning module to use any given pipeline of postprocessors.
  • Refactor shifting and smoothing mechanisms as postprocessors.

agpenas added 10 commits April 13, 2026 18:04
…ce new filtering and shifting classes

- Updated `BaseLabelTuner` to `BasePostprocessor` and added `ForwardLookingFilter` and `Shifter` classes for enhanced label processing.
- Implemented `verify_time_series_and_labels` function for input validation in `BasePostprocessor`.
- Enhanced `RemainingValueTuner` to support a list of postprocessors, allowing for flexible processing pipelines.
- Added comprehensive docstrings and type hints for improved documentation and code quality.
- Introduced `filtering.py` and `shifting.py` modules to encapsulate new filtering and shifting functionalities.
…ardLookingFilter and Shifter post-processor

- Introduced `test_filtering.py` to validate the functionality of `ForwardLookingFilter` with various scenarios, including edge cases and input validation.
- Created `test_shifting.py` to test the `Shifter` class, ensuring correct behavior for forward and backward shifts, as well as handling invalid constructor inputs.
- Updated `test_remaining_value_tuner.py` to support a list of postprocessors in the tuning process, enhancing flexibility in the testing framework.
… postprocessors

- Updated examples in `examples.rst` to include detailed descriptions of optional postprocessors for the `RemainingValueTuner`, such as `ForwardLookingFilter`, `Shifter`, and various smoothing techniques.
- Revised `notebook_examples.rst` to reflect the inclusion of postprocessor pipelines in the label tuning example.
- Expanded `tstrends.label_tuning.rst` to document the new filtering and shifting modules, providing clarity on their usage and functionality.
- Added references to the new modules in the API documentation to improve discoverability and understanding of the label tuning process.
@agpenas agpenas merged commit 100e295 into main Apr 14, 2026
8 of 9 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant