Two Sparse Matrices are Better than One: Sparsifying Neural Networks with Double Sparse Factorization

Open in new window