ELM(n_hidden_neurons::Int,input_data::AbstractArray{T},output_data::AbstractArray{T}; activation::Function = sigmoid, regularization::T = zero(T)) where T<:AbstractFloat

Construct an ELM passing a number of neurons, the inputs and outputs. As a keyword argument, you can also pass a different activation function (Default = sigmoid). Inputs should be in the format below. | Feature1| Feature2| |––––-|––––-| | entry1 | entry1 | | entry2 | entry2 |

predict(elm::ELM,x::AbstractArray{T}) where T<:AbstractFloat

Predict new values