The code is there: http://trac.sagemath.org/ticket/19462
Nathann
On 23 October 2015 at 20:21, Nathann Cohen wrote:
>> What problem are you solving by using immutable matrices/vectors?
>
> I have many vectors v1,...,vk and a ring R, and I want to see if two
> of them are R-collinear (hoping that
> What problem are you solving by using immutable matrices/vectors?
I have many vectors v1,...,vk and a ring R, and I want to see if two
of them are R-collinear (hoping that it's how you guys say it).
> Why is anything immutable for what you are doing?
I multiply vi by R, and take this as a set.
On 23 October 2015 at 20:01, William Stein wrote:
> Hi Nathann,
>
> What problem are you solving by using immutable matrices/vectors? Why
> is anything immutable for what you are doing?
>
> For reference, I explained above the actual problem which motivated my
> writing immutable matrices/vectors
Hi Nathann,
What problem are you solving by using immutable matrices/vectors? Why
is anything immutable for what you are doing?
For reference, I explained above the actual problem which motivated my
writing immutable matrices/vectors for Sage in the first place.
-- William
On Fri, Oct 23, 201
>> I do not see why matrix.row(0) should return a *copy* of the row when
>> the matrix is immutable.
>
>
> It depends on the implementation.
I agree, and that was my point. As you say, under different
circumstances (different implementation of matrices) we may have
chosen to return 'views' on the
On Friday, October 23, 2015 at 6:20:19 AM UTC-7, Nathann Cohen wrote:
>
> I do not see why matrix.row(0) should return a *copy* of the row when
> the matrix is immutable.
>
It depends on the implementation. I think most matrices allocate their
element store as a contiguous block. Hence the vect
> So we agree that there is no performance problem, great. Then don't try to
> illustrate your problem with computational complexity, because as you said
> that is not a problem.
O_o
Nathann
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To uns
On Friday, October 23, 2015 at 3:52:18 PM UTC+2, Nathann Cohen wrote:
>
> Volker, please do not change the subject.
I did not.
> I did not say that complexity was a problem.
So we agree that there is no performance problem, great. Then don't try to
illustrate your problem with computation
> None of the examples you quoted would have a lower complexity with O(1) row
> references. If temporaries are a real problem (Benchmarks first, please)
Volker, please do not change the subject. I did not say that
complexity was a problem. My problem is with the behaviour of .row
which returns mut
On Friday, October 23, 2015 at 3:20:19 PM UTC+2, Nathann Cohen wrote:
>
> Depending on how matrices are implemented matrix.row(0) could very
> well be a O(1) operation
None of the examples you quoted would have a lower complexity with O(1) row
references. If temporaries are a real problem (Benc
> Perhaps should we have a "mutable" argument (True by default to be consistent
> with the current behaviour) in the matrix/vector constructors, so that we can
> do:
>
> {vector(m.row(0), mutable=False) for m in blah}
That would be a "working fix", but really the problem is that
operations
Yo,
> m.row() returns a new (copy) of the row of the matrix, so making that
> immutable would be inconsistent with how copy works for matrices.
>
> sage: m = matrix.ones(10)
> sage: m.set_immutable()
> sage: copy(m).is_immutable()
> False
I do not see why matrix.row(0) should return a *copy* of t
On Fri, Oct 25, 2015 at 06:04:45AM -0700, William Stein wrote:
[...]
>
> At least you can do "v = m.row(0); v.set_immutable()".
There is some issue with this when trying to use set comprehension, see
http://ask.sagemath.org/question/30102/how-do-i-define-and-work-with-a-set-of-matrices/
Perhaps
On Fri, Oct 23, 2015 at 6:04 AM, William Stein wrote:
> On Fri, Oct 23, 2015 at 5:04 AM, Nathann Cohen
> wrote:
>> Hello everybody,
>>
>> When calling .row() on an immutable matrix, one gets mutable vectors.
>> I can easily avoid it for the code I am writing, but do you think we
>> should do som
On Fri, Oct 23, 2015 at 5:04 AM, Nathann Cohen wrote:
> Hello everybody,
>
> When calling .row() on an immutable matrix, one gets mutable vectors.
> I can easily avoid it for the code I am writing, but do you think we
> should do something about it?
m.row() returns a new (copy) of the row of the
Hello everybody,
When calling .row() on an immutable matrix, one gets mutable vectors.
I can easily avoid it for the code I am writing, but do you think we
should do something about it?
sage: m = matrix.ones(10)
sage: m.set_immutable()
sage: hash(m.row(0))
...
TypeError: mutab
16 matches
Mail list logo
