Artificial Neural Network for the Prediction of Tyrosine-Based Sorting Signal Recognition by Adaptor Complexes
Figure 3
ANN training. Connection weights between the Input and hidden layers before (a) and after training (b): values shown were taken from the training of ANN. weights were initialized with small random quantities () as shown in (a) and converged into a broader value range (b). (c) ANN training. The evolution of the total error and accuracy (see main text and Appendix for details) was monitored as a function of the number of Epochs (i.e., number of iterations of a complete set of sequences).
