train_pruning: train a ELM with pruning
In mugiro/elm: Extreme Learning Machine for regression and classification

train a ELM with pruning. The function implements an optimization algorithm for determine the optimium number of neurons

1	train_pruning(object, ...)

`object`	An instance to the ELM class.
`h`	a `matrix` of dimensions [n x l], with the outputs of the hidden layer.
`y`	a `matrix` of dimensions [n x c] or a `vector` if c = 1, with the output values.
`h_val`	a `matrix` of dimensions [n x l], with the outputs of the hidden layer for the validation data.
`y_val`	a `matrix` of dimensions [n x c] or a `vector` if c = 1, with the output values.
`cv_rows`	The vector containing the selection of the data.