“TypeError: 'function' object does not sup

问题:

I have two matrices

fi = [[f1],           Nij = [[N11 N12 .......N1n],
      [f2],                  [N21 N22 .......N2n],
       .                            ...
       .                            ...
      [fn]]                  [Nn1 Nn2 .......Nnn]]

I want to multiply:

f1 to each element of the 1st row,
f2 to each element of the 2nd row,

and so on.

I.e. I want Xij = fi*Nij where fi is a column matrix and Xij & Nij is nxn matrix.

I tried using

import numpy as np

fi = np.linspace(1,5, num =5)
fi = np.asmatrix(fi)

def Xij(ai):
    Nij = np.ones((5,5))
    for i in range(len(fi)):
    for j in range(len(fi)):
        Xij[i,j] = ai[i] * Nij[i,j]
return Xij

Zij = Xij(fi)

It gives me this error TypeError: 'function' object does not support item assignment

Why? and how do I solve this?

回答1:

Your problem is right here:

Xij[i,j] = ai[i] * Nij[i,j]

You named a variable Xij, but also a function. Furthermore, when you named the function, it overwrote the variable.

Because of this, when you try to index the function and assign its items, an error is generated because you can't do that on the function. Below is an example:

>>> def test(): print "Hi"
...
>>> test[1] = "yo!"
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'function' object does not support item assignment
>>>

Does that error look familiar? It is the same one you generated because I did exactly what you did.

To fix this problem, change the name of you function to something other than Xij. Doing so will make Xij equal the matrice, which will support indexing and item assignment.