Table 7 Computational complexity of proposed models.
From: Segmentation-enhanced approach for emotion detection from EEG signals using the fuzzy C-mean and SVM
Model | Total parameters | Approx. FLOPs (millions) |
|---|---|---|
CNN-LSTM hybrid | ~ 380,000 | ~ 70 M |
CNN-LSTM + fuzzy C-means + SVM | ~ 380,000 (CNN-LSTM) + ~ 0 (FCM/SVM) | ~ 70 M (CNN-LSTM only) |
GRU-based flat model | ~ 180,000 | ~ 25 M |
CNN-LSTM + feature extraction + RF | ~ 380,000 | ~ 70 M (CNN-LSTM only) |
