From this mwe:
a=np.zeros((5,5))
b=np.zeros((2,2))
a=np.matrix(a)
b=np.matrix(b)
b[0,0]=4
b[1,1]=9
b[0,1]=7
indice=[2,3]
# 1
c=a[indice,:][:,indice]
c=b
print c
# 2
a[indice,:][:,indice]=b
print a[indice,:][:,indice]
I get:
>>> c
matrix([[ 4., 7.],
[ 0., 9.]])
and:
>>> a[indice,:][:,indice]
matrix([[ 0., 0.],
[ 0., 0.]])
I do not understand why the values of a remain zeroes. If a similar operation is done in two steps, everything works fine:
>>> for k in range(len(indice)):
... a[indice[k],indice]=b[k,:]
I obtain:
>>> a
matrix([[ 0., 0., 0., 0., 0.],
[ 0., 4., 0., 7., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 9., 0.],
[ 0., 0., 0., 0., 0.]])
It's because a[indice,:][:,indice]
isn't a view into the array but is a separate copy -
In [142]: np.may_share_memory(a, a[indice,:][:,indice])
Out[142]: False
To solve it, we can use np.ix_
-
a[np.ix_(indice, indice)] = b
Verify result -
In [145]: a
Out[145]:
matrix([[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.]])
In [146]: b
Out[146]:
matrix([[ 4., 7.],
[ 0., 9.]])
In [147]: a[np.ix_(indice, indice)] = b
In [148]: a
Out[148]:
matrix([[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 4., 7., 0.],
[ 0., 0., 0., 9., 0.],
[ 0., 0., 0., 0., 0.]])
In [149]: a[indice,:][:,indice]
Out[149]:
matrix([[ 4., 7.],
[ 0., 9.]])
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.