Skip to content

Conversation

@carraraig
Copy link
Collaborator

Currently for BNCI2014001, BNCI2014002, BNCI2014004, BNCI2015001

@carraraig carraraig requested a review from bruAristimunha July 29, 2024 14:15
@gcattan
Copy link
Collaborator

gcattan commented Jan 27, 2026

@carraraig Is it the same as in https://arxiv.org/abs/2308.11656 ?
Do you plan to work on it?

@bruAristimunha
Copy link
Collaborator

It is the same @gcattan, but we never merge this because was going against the pipeline logic that we define on how to define the events.

@bruAristimunha
Copy link
Collaborator

@copilot, can you study how to do exact same thing only manipulating the events present within the dataset?

@gcattan
Copy link
Collaborator

gcattan commented Jan 29, 2026

May be could play on tmin, tmax? And run evaluation with different time window.
But also I believe we need to have close implementation to the one of the paper, for reproducibility, if we want to push this on moabb.
wdyt?

@bruAristimunha
Copy link
Collaborator

I feel that we just need to play with tmin and tmax, and change use different split base within the moabb spliter

Resolved conflicts in:
- docs/source/whats_new.rst: Combined Pseudo Online and develop entries
- moabb/datasets/bnci.py: Merged imports (Annotations + find_events)
- moabb/evaluations/utils.py: Kept both _normalized_mcc and _ensure_fitted
- moabb/paradigms/base.py: Added both overlap and scorer parameters
- moabb/paradigms/motor_imagery.py: Merged scoring logic (scorer precedence, then overlap)
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.

4 participants