Activity
From 25 Nov 2019 to 24 Dec 2019
21 Dec 2019
- 17:31 Bug #1376: GBasisTimeout: not working as expected
- John Abbott wrote:
> I have just spent quite a while trying to work out where I could insert a time-out check inside... - 17:22 Bug #1376: GBasisTimeout: not working as expected
- I have just spent quite a while trying to work out where I could insert a time-out check inside the reduction code.
...
20 Dec 2019
- 17:10 Design #1378 (New): Create two separate radical fns (for 0-dim ideals)
- Proposal: create two functions for computing radicals
* *@radical@* which also tries hard to compute a GB for the r... - 17:05 Slug #1375: Radical 0-dim: varied timings
- I propose doing an experiment along the following lines.
We create a collection of test ideals (perhaps as many a... - 16:59 Design #1377: CpuTimeLimit: limit "intervals" between full checks
- I have checked in the code (even though it is not quite finished yet). It worked tolerably well in my two tests.
- 16:34 Design #1377: CpuTimeLimit: limit "intervals" between full checks
- I have an improved impl, which I hope to check in fairly soon (later today?).
A problem with the description in co... - 16:58 Bug #1376: GBasisTimeout: not working as expected
- The main problem with checking for timeout during GB computation is that the times for each single reduction can vary...
14 Dec 2019
- 12:37 Design #1377: CpuTimeLimit: limit "intervals" between full checks
- Now I'm thinking it may be simpler to use an approach like this:
measure actual @CpuTime@ difference over the last... - 12:32 Bug #1376: GBasisTimeout: not working as expected
- Reply to comment 6:
yes, @CpuTimeLimit@ does already attempt to adapt automatically. I'm discussing (with myself?) ...
13 Dec 2019
- 18:17 Design #1377 (In Progress): CpuTimeLimit: limit "intervals" between full checks
- The experimental results above suggest that approach *(B)* is probably too simplistic if there is any risk of a @CpuT...
- 14:33 Design #1377: CpuTimeLimit: limit "intervals" between full checks
- Here are the results of an extreme speed test for option *(B)* in the previous comment.
The simplistic test progra... - 13:46 Design #1377: CpuTimeLimit: limit "intervals" between full checks
- My current suggestions:
* *(A)* let the creator of the @CpuTimeLimit@ object specify a maximum interval between genu... - 13:39 Design #1377: CpuTimeLimit: limit "intervals" between full checks
- In the current CVS version, intervals between genuine checks can be as long as 1000000 calls.
*Quick reminder*: is... - 13:30 Design #1377 (Closed): CpuTimeLimit: limit "intervals" between full checks
- A *@CpuTimeLimit@* object actually genuinely checks the time limit only every @N@ calls to the check function. The v...
- 13:26 Bug #1376: GBasisTimeout: not working as expected
- I was thinking of an alternative approach since the timeout mechanism is applied in so different cases:
The interval... - 12:45 Bug #1376: GBasisTimeout: not working as expected
- An important point to note is that the design of @CpuTimeLimit@ works best when the times between successive calls to...
- 11:57 Slug #1375: Radical 0-dim: varied timings
- Instead of guessing, do this ;-)...
12 Dec 2019
- 21:44 Bug #1376 (In Progress): GBasisTimeout: not working as expected
- It seems that my guess was good.
In @CpuTimeLimit.C@ around line 67, the code allowed the interval between full ch... - 17:04 Bug #1376: GBasisTimeout: not working as expected
- I noticed the problem while using a simple program to look for ideals where it was costly to compute the RGB of the r...
- 17:02 Bug #1376 (Closed): GBasisTimeout: not working as expected
- It seems that sometimes @GBasisTimeout@ does not timeout as one would expect.
I have just interrupted (via C-c C-c... - 15:30 Slug #1375: Radical 0-dim: varied timings
- I have modified my program to call @GBasisTimeout@, and I specified 30s as the time-out.
I was a bit surprised to fi... - 13:33 Slug #1375: Radical 0-dim: varied timings
- Here are some more examples where it takes a long time to compute RGB after computing radical:...
- 11:33 Slug #1375: Radical 0-dim: varied timings
- The current solution (increasing the "increment" to 0.5) works well on the small examples I have tried: sometimes the...
11 Dec 2019
- 15:08 Slug #1375 (In Progress): Radical 0-dim: varied timings
- Following Anna's suggestion I moved @t0 := CpuTime();@ to just before the computation of the radical; so the time dif...
10 Dec 2019
- 15:12 Slug #1375: Radical 0-dim: varied timings
- Sometimes radical is computed with the GBasis, sometimes not: this depends on a timeout, which probably is a very cri...
- 14:41 Slug #1375: Radical 0-dim: varied timings
- I have done more testing. The timings really do vary from run to run.
I wrote a shell script which took exactly t... - 13:28 Slug #1375: Radical 0-dim: varied timings
- This getting more puzzling!
Here is another example: essentially the same code as in the original description but
...
09 Dec 2019
- 21:08 Slug #1375: Radical 0-dim: varied timings
- I wanted to show the students an example where it is much better to use @IsInRadical@ than compute the radical (with ...
- 21:06 Slug #1375 (Closed): Radical 0-dim: varied timings
- I have observed some strange variations in timings when computing the RGB of a radical:...
- 13:34 Slug #1136: IsInRadical: sometimes a bit slow
- Just for the record...
With the exponent pair (3,4), my computer takes about 2.4s whereas with the pair (4,5) it t... - 12:56 Slug #1136 (Closed): IsInRadical: sometimes a bit slow
- Instant on my computer. I wonder how it was improved....
05 Dec 2019
- 15:55 Slug #1136 (Feedback): IsInRadical: sometimes a bit slow
- It took about 0.08s on my computer just now. This is probably fine; I see no advantage in making it lengthier.
Move... - 15:52 Slug #1136: IsInRadical: sometimes a bit slow
- What is the status of this issue?
Can it be closed soon?
04 Dec 2019
- 17:38 Bug #1233: factor SEGV -- too large characteristic
- This now gives error rather than SEGV -- so that is an improvement!
- 17:19 Support #1161: Split NumTheory.C into several smaller files
- Postponing because not important for CoCoA School.
27 Nov 2019
- 15:53 Bug #1371: French students' example with GFan
- I find it quite strange that the term ordering used to obtain the same LT(I) actually looks to be far away from lex:
...
26 Nov 2019
- 15:45 Bug #1371: French students' example with GFan
- Just for curiosity, this lex GBasis can be computed instantly using *GBasisByHomog(I)*.
Then, together with Robbia... - 15:10 Bug #1371 (In Progress): French students' example with GFan
- The time taken depends on the current ring ordering!
If I create the ideal in the ring @QQ[x,y,z]@ with DegRevLex,...
25 Nov 2019
- 20:33 Slug #1369 (In Progress): RandomUnimodularMat is slow with many iters
- First impl is quite disappointing; the overhead/cost of matrix mult is embarrassingly high.
Should do recursion base... - 13:48 Slug #1369: RandomUnimodularMat is slow with many iters
- @RandomUnimodularMat(QQ,4,1000000)@ takes too long. It is faster to do @product([RandomUnimodularMat(QQ,4,10000) | j...
- 13:45 Slug #1369 (In Progress): RandomUnimodularMat is slow with many iters
- @RandomUnimodularMat@ becomes too slow when number of iters is high.
Try a divide-and-conquer approach. - 17:46 Bug #1371: French students' example with GFan
- Just for completeness here is the input:...
- 17:44 Bug #1371 (In Progress): French students' example with GFan
- I have just tried the French students' example as argument to @GroebnerFanIdeals@, and it computed all 167 GBases qui...
- 16:35 Design #1370: RingID counter increased inside some functions
- John Abbott wrote:
> We could perhaps have two counters: one for "user created" rings and one for "internal" rings. ... - 14:35 Design #1370: RingID counter increased inside some functions
- We could perhaps have two counters: one for "user created" rings and one for "internal" rings. Or perhaps just posit...
- 14:32 Design #1370 (New): RingID counter increased inside some functions
- Each ring in CoCoALib has a unique ring-ID (machine integer).
Some functions (_e.g._ *@syz@*) create new rings int...
Also available in: Atom