代码如下:
from scipy import sparse
from scipy.sparse import coo_matrix
from scipy.sparse import vstack
import numpy as np
data = [5,2,3,0]
row = [2,2,3,2]
col = [3,4,2,3]
c = [coo_matrix((data,(row,col)),shape=(5,6))]*3
ccc=vstack(c)
print(type(ccc))
sparse.save_npz("ccc.npz",ccc)
ddd=sparse.load_npz("ccc.npz")
print(type(ddd))
如图所示:
稀疏矩阵保存与加载都有scipy.sparse内置模块完成;
实验中,我也np.save的方法,但加载后不再是稀疏矩阵格式