Research Article
Semisupervised Seizure Prediction in Scalp EEG Using Consistency Regularization
Require: = STFT input, . | Require: = ground truth only for labeled input, . | Require: = stochastic Gaussian augmentation function. | Require: = the network with trainable parameters . | Require: = ensembling constant, . | Require: = ramp-up weighting function. | (1) | Initialize the ensemble prediction . | (2) | Initialize the target vector . | (3) | For in do | (4) | For each minibatch do | (5) | . | (6) | . | (7) | . | (8) | . | (9) | Update using Adam optimizer | (10) | end for | (11) | . | (12) | . | (13) | end for | (14) | Return . |
|