Fig. 4From: High-dimensional neural feature design for layer-wise reduction of training costTraining and testing accuracy against size of HNF using ELM feature vector in the first layer and DCT in the next layers. Size of an L-layer HNF is represented by the number of nodes, counted as \(n^{(1)} + \sum _{l=2}^{L} 2n^{(l)}\). Here, L=3 for all three datasets. The number of nodes in the first layer (n(1)) is set according to Table 2 for each datasetBack to article page