Fig. 8
From: Low-abundant bacteria drive compositional changes in the gut microbiota after dietary alteration
Topology of the ANN used to train on sequenced data. The number of input nodes was set to the number of taxonomic orders (T) plus the number of substrates (S) (70 total). The number of nodes in the first hidden layer was set to 95% of the total input nodes, whereas the number of nodes in the second hidden layer was set to the 85%. The number of output nodes was set to the number of taxonomic orders, as the goal of the network was to predict relative abundance changes over time for each taxon. The arrangement of taxonomic orders remained constant for each CSV file