Time series forecasting with an intuitionistic fuzzy artificial
neural network. Lagged values of the series and their intuitionistic
fuzzy membership and non-membership degrees, obtained by intuitionistic
fuzzy c-means clustering, are used as inputs of a feed-forward neural
network with one hidden layer, averaged over several random starts. The
lag order and the network hyper-parameters (number of hidden neurons and
weight decay) can be set manually or tuned automatically by grid search
over user-defined minimum, maximum and step ranges, with time-ordered
validation and a progress display. The intuitionistic fuzzy approach
follows Yolcu and Yolcu (2023)