This commit is contained in:
Michael Weig 2026-03-10 13:11:03 +01:00
parent 910e642398
commit 9c2619daa9

View File

@ -27,11 +27,12 @@ Main scripts:
- `dataset_creation/parquet_file_creation.py` - `dataset_creation/parquet_file_creation.py`
Purpose: Purpose:
- Read source recordings (`.h5` and/or ownCloud-fetched files) - Download and/or access dataset files (either download first via ```EDA/owncloud_file_access.ipynb``` or all in one with ```dataset_creation/create_parquet_files_from_owncloud.py```
- Keep relevant simulator/physiology columns - Keep relevant columns (FACE_AUs and eye-tracking raw values)
- Filter invalid samples (e.g., invalid level segments) - Filter invalid samples (e.g., invalid level segments): Make sure not to drop rows where NaN is necessary for later feature creation, therefore use subset argument in dropNa()!
- Export subject-level parquet files - Export subject-level parquet files
### 2.2 Feature Engineering (Offline) ### 2.2 Feature Engineering (Offline)
Main script: Main script: