Storage processing method and device for sparse matrix

The invention relates to a storage processing method and device for a sparse matrix, and the method comprises the steps: dividing the sparse matrix into N sub-matrixes, and determining the line numbers and column numbers corresponding to the N sub-matrixes; and, obtaining non-zero elements from the sub-matrixes in sequence to form a non-zero element array; obtaining bit sequences respectively corresponding to the N sub-matrixes, wherein for any ith sub-matrix in the N sub-matrixes, when the jth element in the ith matrix is a zero element, the value of the jth bit in the ith bit sequence corresponding to the jth element is 0; when the jth element in the ith matrix is a non-zero element, the value of the jth bit in the ith bit sequence corresponding to the jth element is 1; and sequentially storing the row numbers, the column numbers, the bit sequences and the non-zero element arrays corresponding to the N sub-matrixes respectively. By adopting the method provided by the embodiment of the invention, a large amo