CoCoA » CoCoALib

18 May 2016

13:44 Slug #881 (Resolved): ReadExpr is too slow on large polys

I have modified @ReadExprInSparsePolyRing@ so that sums of terms are handled specially: the terms are simply appended... John Abbott

11 May 2016

14:48 Slug #881: ReadExpr is too slow on large polys

I reran the test to read Mario's polynomial: now it takes about 0.4s (with the "wrong" term-ordering), and about 0.25... John Abbott

14:32 Slug #881: ReadExpr is too slow on large polys

I have now finished the main changes to @DistrMPolyInlPP@, and especially @myAddClear@. Time to read a correctly ord... John Abbott

12:09 Slug #881: ReadExpr is too slow on large polys

I wrote a very hacked version which uses @PushBack@, and it went considerably faster (but it also gave SEGV on some e... John Abbott

10 May 2016

16:36 Slug #881: ReadExpr is too slow on large polys

I have already implemented (and checked in) a version which uses geobuckets when the ring is a sparsepolyring.
Mar... John Abbott

16:19 Slug #881: ReadExpr is too slow on large polys

John Abbott wrote:
> I have done a quick profile of the current version of the code (using the _ad hoc_ impl with @s... Anna Maria Bigatti

15:33 Slug #881 (In Progress): ReadExpr is too slow on large polys

I have done a quick profile of the current version of the code (using the _ad hoc_ impl with @std::map@). Now most (... John Abbott

13:25 Slug #881: ReadExpr is too slow on large polys

I have hacked the @ReadExpr@ code so that it produces a table of symbols (of type @std::map@).
This improved the rea... John Abbott

09 May 2016

22:10 Slug #882: Impl faster multiplication for DUP (dense univariate polys)

I also tried implementing a method which uses fast integer multiplication in GMP.
Evaluate both polys at 1000000 (... John Abbott

22:07 Slug #882: Impl faster multiplication for DUP (dense univariate polys)

I made a very _ad hoc_ impl of karatsuba when I was in Osnabrueck as I wanted to get an idea of when cross-over might... John Abbott

22:05 Slug #882 (New): Impl faster multiplication for DUP (dense univariate polys)

Christof wanted faster multiplication for DUPs.
He suggested implementing karatsuba's method. John Abbott

21:49 Slug #881: ReadExpr is too slow on large polys

I did a quick test:... John Abbott

21:23 Slug #881: ReadExpr is too slow on large polys

Since my "stupid" MacBook cannot profile, I can only guess that reading a symbol is probably quite costly (see line 7... John Abbott

21:13 Slug #881: ReadExpr is too slow on large polys

I have implemented a first improvement by implementing a new fn @ReadExprInSparsePolyRing@ which uses geobuckets to s... John Abbott

21:10 Slug #881 (Closed): ReadExpr is too slow on large polys

Recently Mario has passed me some large polys in their printed form (_i.e._ sum of terms). Reading them by pasting t... John Abbott

20:51 Slug #874: factor: too slow on largish multivariate poly

I have implemented a prototype "heuristic" irred test which returns a @bool3@.
First it checks for a non-triv comm... John Abbott

06 May 2016

16:43 Slug #866: implicit, ImplicitHypersurface: improve output verification

I have not yet had time to look at your horner code. Anyway, I would expect it work fairly well provided the "variab... John Abbott

03 May 2016

13:14 Slug #874: factor: too slow on largish multivariate poly

Mario has just sent me an even bigger polynomial to try: 100 indets, about 127000 terms, total degree 9.
He also m... John Abbott

10:50 Slug #874 (In Progress): factor: too slow on largish multivariate poly

Computing all contents takes about 44s; and computing contents wrt all indets which appear just linearly takes about ... John Abbott

10:42 Slug #874: factor: too slow on largish multivariate poly

Mario has given me an example poly. The poly ring has 84 indets; the poly itself uses 69 indets, and has about 6800 ... John Abbott

10:37 Slug #874 (In Progress): factor: too slow on largish multivariate poly

Mario generates lots of multivariate polys (over @QQ@); most of them are irreducible. He reports that factorization ... John Abbott

10:34 Support #861 (In Progress): Janet basis code: TmpJB files give some problems with C++11 (using CLANG/LLVM)

Mario has given me an updated version of his code. He has used the @__cplusplus@ preprocessor variable to have code ... John Abbott

29 Apr 2016

22:00 Feature #873: bool3: add some logical operations?

(Maybe this is what you had in mind from the beginning)
I think the only safe definition for @and3@ is the @and@ bet... Anna Maria Bigatti

19:21 Feature #873: bool3: add some logical operations?

The "monomial generators" case is easy.
The "is a monomial ideal" is hard in any case: even if both ideals are not m... John Abbott

18:53 Feature #873: bool3: add some logical operations?

I expect in some cases we want
uncertain3 and false3 = false3 (sum of ideals has monomial generators)
and other t... Anna Maria Bigatti

18:15 Feature #873 (In Progress): bool3: add some logical operations?

It would not be harmful to your code if @bool3@ offered the functions, but you may be right that it would not simplif... John Abbott

28 Apr 2016

10:45 Feature #873: bool3: add some logical operations?

I'm against this idea. Indeed it might be handy, but dangerous as well: while working on fixing a bug on the bool3 f... Anna Maria Bigatti

27 Apr 2016

18:49 Feature #873: bool3: add some logical operations?

BOOST also offers a "not" operation. It would be trivial to implement @not3@ (giving value 1 - arg).
If we do off... John Abbott

18:45 Feature #873: bool3: add some logical operations?

Part of the motivation comes from Anna's rectification of code for ideal arithmetic (see issue #870).
For example ... John Abbott

18:42 Feature #873: bool3: add some logical operations?

I suggest adding the following two operations:
* *and* which is equivalent to "minimum"
* *or* which is equivalent... John Abbott

18:37 Feature #873 (In Progress): bool3: add some logical operations?

Currently there are no logical operations on @bool3@ values.
Discuss whether it may be helpful to include some sim... John Abbott

26 Apr 2016

19:05 Feature #872 (In Progress): LinSolve for matrices over FFp

Currently CoCoALib uses generic code for solving linear systems over small finite fields.
Implement some special c... John Abbott

15:01 Design #871 (New): Redesign ideals

Review the design of ideals.
Some points to bear in mind:
* an ideal should probably be immutable -- any operatio... John Abbott

22 Apr 2016

18:09 Slug #866: implicit, ImplicitHypersurface: improve output verification

I checked in @IsZeroEvalHorner@ in case you want to play with it.
The call is currently commented out. Just look fo... Anna Maria Bigatti

18:01 Slug #866: implicit, ImplicitHypersurface: improve output verification

Anna Maria Bigatti wrote:
> > But I'd like another try: I'm planning to try to implement a sort of recursive Horner.... Anna Maria Bigatti

17:02 Slug #866: implicit, ImplicitHypersurface: improve output verification

I seem to recall vaguely seeing a paper by Mourrain (or someone in his group) possibly about this. I think the quest... John Abbott

16:26 Slug #866: implicit, ImplicitHypersurface: improve output verification

Anna Maria Bigatti wrote:
> But I'd like another try: I'm planning to try to implement a sort of recursive Horner. M... Anna Maria Bigatti

15:14 Feature #869 (New): SmallFpImpl: different repr for computation and for storage?

I would like to use @unsigned short@ to store @SmallFp@ values, but still have a decent range _i.e._ without the rest... John Abbott

19 Apr 2016

14:59 Bug #867: Compilation Error: C++11 and old GMP-Version

Perhaps I should have read the @configure@ script better: I have already inserted (ages ago) a warning for versions o... John Abbott

14:54 Bug #867: Compilation Error: C++11 and old GMP-Version

John Abbott wrote:
> Mario suggested that we could modify the CoCoALib configuration script to check that GMP is at ... Anna Maria Bigatti

14:20 Bug #867 (In Progress): Compilation Error: C++11 and old GMP-Version

Mario suggested that we could modify the CoCoALib configuration script to check that GMP is at least version 5.0.0; c... John Abbott

14:27 Feature #82 (In Progress): C++11 compatibility questions

Mario has found a web page which explains that the preprocessor flag @__cpluscplus@ has a numerical value indicating ... John Abbott