#19666: Guruswami-Sudan decoder for GRS codes
-------------------------------------+-------------------------------------
Reporter: dlucas | Owner:
Type: enhancement | Status: needs_review
Priority: major | Milestone: sage-7.1
Component: coding theory | Resolution:
Keywords: | Merged in:
Authors: Johan Sebastian | Reviewers: dlucas
Rosenkilde Nielsen, David Lucas | Work issues:
Report Upstream: N/A | Commit:
Branch: | 2b0975090d03abeef343811a934646a33bcedc16
u/jsrn/gs_list_decoding | Stopgaps:
Dependencies: #18928 |
-------------------------------------+-------------------------------------
Comment (by dlucas):
I read all your changes, and I agree with most, I made a few small changes
by myself:
- There was an typo in `decode_to_message`'s doctest because of which the
test failed. I fixed it.
- `_flatten_once` doctest had the flag `#random` while its output is not
random, thus I removed this flag.
One remark: In #19653, I changed all my hardcoded doctests related to
`decode_to_*` methods by examples looking like this:
{{{
build the code, say C
build the decoder, say D
generate a random element of the code, say c
create a channel which creates as many errors as the decoder's decoding
radius
transmit c to get y
check if c == D.decode_to_code(y) [resp. D.connected_encoder().unencode(c)
== D.decode_to_message(y)]
}}}
which I think is a bit better. For instance, it allows us to pick larger
codes for tests as you don't have to write the codewords...
Here, I noticed you picked your examples so you get a list of `size > 1`,
which I think is a good idea. I don't see a way to use this kind of tests
AND guarantee that list size will be at least two though...
Also, I don't really like the new title of `interpolation.py` module as it
appears in the index of modules:
> Finding F[x]-roots, or modular F[x] roots, in polynomials over F[x][y],
where F is a (finite) field.
With such a title, one does not immediately see how this module refers to
codes/decoding imho.
It's a matter of taste I guess, so I'm not asking to change it, I'm just
pointing it.
On my side, I give the greenlight to all your changes. If you agree with
mine, I think we're good to go.
David
--
Ticket URL: <http://trac.sagemath.org/ticket/19666#comment:28>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/d/optout.
