Compute digest from the full processing+classification pipeline #429
Assignees
Labels
Comments
This was referenced Jul 26, 2023
Merged
PierreGtch
added a commit
to PierreGtch/moabb
that referenced
this issue
Aug 3, 2023
sylvchev
pushed a commit
that referenced
this issue
Aug 25, 2023
* Add methods get_process_pipelines and get_labels_pipeline to paradigms * Compute digest from full pipeline Closes #429 * Update docstrings * Update whats_new.rst * Add process_pipeline parameter to Results.not_yet_computed * Add process_pipeline parameter to Results.to_dataframe() * Update benchmark test * Fix prepare_process not called when make_process_pipelines called aoutside of get_data * Complete TODOs * Add shortcut for make_process_pipelines in utils * [pre-commit.ci] auto fixes from pre-commit.com hooks * Remove TODO as discussed (only keep pipeline for first band) --------- Co-authored-by: Bru <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
With PR #408 and commit 8c88349, the data processing steps (filtering, resampling, epoching...) are done by a sklearn pipeline.
So we could use the full pipeline (procesing+classification) to create the code that identifies the pipeline in the results, instead of only the classifier.
This would allow to differentiate pipelines with the same classifier but different pre-processing steps.
The text was updated successfully, but these errors were encountered: