Talk:Diophantine linear system solving: Difference between revisions
No edit summary |
(→Totally lost: slowly getting a few bits) |
||
Line 38: | Line 38: | ||
But for now: if more people find this task...objectionable, it probably doesn't belong on RC after all. |
But for now: if more people find this task...objectionable, it probably doesn't belong on RC after all. |
||
--M.C., [[User:Udo_e._pops|Udo e.]] ([[User talk:Udo e. pops|talk]]) |
--M.C., [[User:Udo_e._pops|Udo e.]] ([[User talk:Udo e. pops|talk]]) |
||
Thanks, I have slowly figured a couple of things out. That example means, I think, solve {2a+b=2, 6a+5b=2, 7a+6b=2} ==> the answer, from (-2 2)*-1, is {a=2,b=-2}. I was thinking (and about to say) that a proper set of unit tests on Reduce() would help, but actually what I think I really need right now is unit tests (/detailed examples) on Swop(), especially with what it is trying to do to d[], so I can properly visualise (and maybe even understand) it. --[[User:Petelomax|Pete Lomax]] ([[User talk:Petelomax|talk]]) 16:40, 25 December 2021 (UTC) |
Revision as of 16:41, 25 December 2021
Totally lost
Some questions:
What does any of this actually mean?
What is a basis?
What is the standard basis?
What is a vector space?
What is the dimension of the vector space?
What is linear independence?
What is spanning property?
What is a spanning set?
What is a polynomial ring?
What is a monomial?
What is a monomial basis?
What is the Steinitz exchange lemma?
What is the axiom of choice?
What is the ultrafilter lemma?
What is the dimension theory?
I think you can see where I'm going with this, and I'm not even halfway through the first of, I dunno, 30,000 wikipedia pages I would have to read.
Let's try this: what does this actually mean?
2 1| 2 6 5| 2 7 6| 2
==>
P | Hnf 1 -1 0 | 1 1 1 0 -1 2 -0 | -0 4 5 -0 -2 2 1 | 0 0 0 1 -solution loop 7
I have a very vague grasp of what the input is, but the output is pure gibberish, and obviously "Transformation matrix P (left) and the Hermite normal form" elicits qs as above.
Detailed step-by-step inner workings of all 21 puzzles might be a start... --Pete Lomax (talk) 22:24, 24 December 2021 (UTC)
Sorry Pete, this task wasn't intended to swamp you in some murky math morass.
Let me reflect on this. I have a few busy days ahead, will be back sometime after Xmas.
But for now: if more people find this task...objectionable, it probably doesn't belong on RC after all.
--M.C., Udo e. (talk)
Thanks, I have slowly figured a couple of things out. That example means, I think, solve {2a+b=2, 6a+5b=2, 7a+6b=2} ==> the answer, from (-2 2)*-1, is {a=2,b=-2}. I was thinking (and about to say) that a proper set of unit tests on Reduce() would help, but actually what I think I really need right now is unit tests (/detailed examples) on Swop(), especially with what it is trying to do to d[], so I can properly visualise (and maybe even understand) it. --Pete Lomax (talk) 16:40, 25 December 2021 (UTC)