the data to feed to the roots in proper order
array of last layer neurons values
// net with a single DenseData(2) root: net.predict([3.2f, -1.5f]); // net with a single SparseData(dim >= 34) root: net.predict([SparseF(34, -0.7f), SparseF(3, 0.2f)]); // net with one DenseData(1), one SparseData(dim >= 16) root: net.predict([0.2f], [SparseF(16, -0.15f)]);
Compute the prediction of the net when passing the arguments to the root(s) of the net.