deepak2233/AMIGOS

Affective computing with AMIGOS dataset

Requirement

Python == 3.6.3
biosppy == 0.5.0
EMD-signal == 0.2.3
numpy == 1.13.3
scipy == 0.19.1
scikit-learn == 0.19.0
xgboost == 0.6

Usage

# Preprocessing(features stored in data/features.p)
python preprocess.py --input [AMIGOS_DATA_DIRECTORY (default is ./data)]

# Training and Cross Validation
python main.py --data [AMIGOS_DATA_DIRECTORY (default is ./data)]
               --feat [MODALITY_TYPE (default is all)]
               --clf [CLASSIFIER_TYPE (default is xgb)]
               --nor [NORMALIZATION_METHOD (default is no)]

# Print help message
python main.py -h