Fig. 6From: A comparative study of CNN-capsule-net, CNN-transformer encoder, and Traditional machine learning algorithms to classify epileptic seizureTransformer Encoder. Transformer encoder input is the output of the CNN. The figure contains Multi-Head Attention layers that weigh the relevance of each input vector. The MLP processes these representations of each vector. Residual Connections and Layer Normalization facilitate an efficient and stable flow of information. The output is a deep contextual representation of the inputBack to article page