In [1]:

    

from scipy.sparse import csr_matrix, csc_matrix, coo_matrix


    

In [2]:

    

data = [10, 20, 30, 40]
row = [0, 0, 1, 1]
col = [1, 2, 0, 2]


    

In [3]:

    

print(csr_matrix((data, (row, col))).toarray())


    

    




[[ 0 10 20]
 [30  0 40]]

In [4]:

    

print(csr_matrix((data, (row, col)), shape=(3, 3)).toarray())


    

    




[[ 0 10 20]
 [30  0 40]
 [ 0  0  0]]

In [5]:

    

print(csr_matrix((data, (row, col)), shape=(4, 4)).toarray())


    

    




[[ 0 10 20  0]
 [30  0 40  0]
 [ 0  0  0  0]
 [ 0  0  0  0]]

In [6]:

    

data = [10, 20, 30, 40]
row = [0, 0, 1, 1]
col = [1, 2, 2, 2]


    

In [7]:

    

print(csr_matrix((data, (row, col))))


    

    




  (0, 1)	10
  (0, 2)	20
  (1, 2)	70

In [8]:

    

print(csc_matrix((data, (row, col))))


    

    




  (0, 1)	10
  (0, 2)	20
  (1, 2)	70

In [9]:

    

print(coo_matrix((data, (row, col))))


    

    




  (0, 1)	10
  (0, 2)	20
  (1, 2)	30
  (1, 2)	40

In [10]:

    

print(coo_matrix((data, (row, col))).tocsr())


    

    




  (0, 1)	10
  (0, 2)	20
  (1, 2)	70

In [11]:

    

print(coo_matrix((data, (row, col))).toarray())


    

    




[[ 0 10 20]
 [ 0  0 70]]

In [12]:

    

print(csr_matrix(([10, 20, 30, 40], [1, 2, 0, 2], [0, 2, 4, 4]), shape=(3, 3)).toarray())


    

    




[[ 0 10 20]
 [30  0 40]
 [ 0  0  0]]

In [13]:

    

print(csc_matrix(([30, 10, 20, 40], [1, 0, 0, 1], [0, 1, 2, 4]), shape=(3, 3)).toarray())


    

    




[[ 0 10 20]
 [30  0 40]
 [ 0  0  0]]