AttributeError: 'int' object has no attribute 'inverse_mod'

inverse_mod

asked 2016-06-04 14:46:45 +0100

fagui
187 ●10 ●18 ●24

updated 2016-06-05 12:28:10 +0100

slelievre
17704 ●22 ●160 ●350 http://carva.org/samue...

Hi i'm a beginner with Sage, and i have some trouble with types. It looks like the cause of my pain is trying to use numpy together with Sagemath, is it possible to use Sagemath without numpy?

Here is a small example of an error that i encounter, when trying to do simple modular calculations in GF(7).

I use the following code:

import numpy as np

a=[2,4,6,1,3,5]
b=[1,1,1,1,1,1]
p=7
n=len(a)
L=1
c=np.zeros(n,'int')
Li_ai=np.zeros(n)
for i in range(n):
    L=(x-a[i])*L
for i in range(n):
    Li=L/(x-a[i])
    t=(Li(x=a[i])%p)
    Li_ai[i]=t
Li_ai=Li_ai.astype('int')
tmp=Li_ai.prod()%p
Li_ai=vector(Li_ai)
for i in range(n):
    c[i]=tmp * (Li_ai[i]) * b[i].inverse_mod(p) %p
c=vector(c)
c
y=np.int(tmp)
y.inverse_mod(p)

and I get the following error:

Traceback (most recent call last)
...
AttributeError: 'int' object has no attribute 'inverse_mod'

edit retag flag offensive close merge delete

add a comment

answered 2016-06-04 14:53:22 +0100

slelievre
17704 ●22 ●160 ●350 http://carva.org/samue...

updated 2016-06-05 12:42:01 +0100

(Edited to answer the follow-up question posted as a comment.)

Your code uses numpy, which uses Python's int type.

Sage has its own type of integers, in addition to Python integers.

Sage methods such as inverse_mod are implemented for Sage integers, not Python integers.

You can convert a Python integer k to a Sage integer by calling Integer(k) or ZZ(k).

So, if you define, as in your code,

y = np.int(tmp)

then change the last line to

ZZ(y).inverse_mod(p)

Better: directly define

y = ZZ(tmp)

and then do

y.inverse_mod(p)

Now, you might want to avoid numpy altogether and work with native Sage matrices and vectors.

You can initialize a matrix or vector by specifying the base ring (eg ZZ, QQ, etc), the dimension, and the entries. If you don't provide entries, you get a zero matrix or vector. Matrices and vectors are mutable by default.

Rows, columns, and vector components are indexed from 0 rather than 1.

sage: a = matrix(ZZ, 3)
sage: a
[0 0 0]
[0 0 0]
[0 0 0]
sage: a[0, 0] = 4
sage: a
[4 0 0]
[0 0 0]
[0 0 0]

sage: v = vector(ZZ, 3)
sage: v
(0, 0, 0)
sage: v[1] = 2
sage: v
(0, 2, 0)
sage: a * v
(0, 0, 0)

You can find lots more information and examples in the documentation and tutorials, eg

http://doc.sagemath.org/html/en/tutorial/tour_linalg.html