Code for analyzing keypress collected during the EEG study. Protocol discussion: Choose or Fuse: Enriching Data Views with Multi-label Emotion Dynamics. Cang et al. (ACII 2022)
Currently converting from Jupyter nb to python module, not ready for usage.
read_data.py
: utils to load CSV filesclean_data.py
: utils to clean up raw files, i.e., fix scene and keystroke flags, fix sampling to 30H, create keys a5 (sum of all keypress values) and a6 (max of all keypress values)calculate_features.py
: utils to calculate features from keypress data (statistical, frequency and keystroke features)config.py
: constant declarationstrain.py
: main training script, training pipeline defined inestimator_helper.py
(grid search cross-validation with recursive feature elimination)utils.py
: general purpose methods (pickle and load pickled files)
- Debug
- Optimize methods
- Plot results
-
create_training_dataset.py