Newest at the top
| 2025-03-24 22:01:45 +0100 | ash3en | (~Thunderbi@89.56.182.235) (Client Quit) |
| 2025-03-24 22:00:37 +0100 | kh0d | (~kh0d@212.200.65.82) |
| 2025-03-24 22:00:31 +0100 | ash3en | (~Thunderbi@89.56.182.235) ash3en |
| 2025-03-24 21:59:23 +0100 | alfiee | (~alfiee@user/alfiee) (Ping timeout: 244 seconds) |
| 2025-03-24 21:55:34 +0100 | merijn | (~merijn@host-vr.cgnat-g.v4.dfn.nl) (Ping timeout: 272 seconds) |
| 2025-03-24 21:55:05 +0100 | alfiee | (~alfiee@user/alfiee) alfiee |
| 2025-03-24 21:53:53 +0100 | kh0d | (~kh0d@212.200.65.82) (Remote host closed the connection) |
| 2025-03-24 21:52:02 +0100 | kh0d | (~kh0d@212.200.65.82) kh0d |
| 2025-03-24 21:51:44 +0100 | jespada | (~jespada@2800:a4:22bd:1300:6874:df1d:56c8:ac1f) (Quit: My Mac has gone to sleep. ZZZzzz…) |
| 2025-03-24 21:50:07 +0100 | merijn | (~merijn@host-vr.cgnat-g.v4.dfn.nl) merijn |
| 2025-03-24 21:50:00 +0100 | weary-traveler | (~user@user/user363627) (Remote host closed the connection) |
| 2025-03-24 21:45:42 +0100 | chele | (~chele@user/chele) (Remote host closed the connection) |
| 2025-03-24 21:44:17 +0100 | kh0d | (~kh0d@212.200.65.82) (Remote host closed the connection) |
| 2025-03-24 21:37:32 +0100 | peterbecich | (~Thunderbi@syn-047-229-123-186.res.spectrum.com) peterbecich |
| 2025-03-24 21:35:54 +0100 | ThePenguin | (~ThePengui@cust-95-80-24-166.csbnet.se) ThePenguin |
| 2025-03-24 21:33:55 +0100 | ThePenguin | (~ThePengui@cust-95-80-24-166.csbnet.se) (Remote host closed the connection) |
| 2025-03-24 21:31:52 +0100 | Tuplanolla | (~Tuplanoll@91-159-69-59.elisa-laajakaista.fi) Tuplanolla |
| 2025-03-24 21:27:45 +0100 | alp | (~alp@2001:861:8ca0:4940:92ce:4d9a:3c9b:8560) (Ping timeout: 246 seconds) |
| 2025-03-24 21:22:45 +0100 | acidjnk | (~acidjnk@p200300d6e71c4f84e5bb842294cfb41d.dip0.t-ipconnect.de) (Ping timeout: 248 seconds) |
| 2025-03-24 21:22:31 +0100 | wootehfoot | (~wootehfoo@user/wootehfoot) (Read error: Connection reset by peer) |
| 2025-03-24 21:20:06 +0100 | CiaoSen | (~Jura@2a02:8071:64e1:da0:5a47:caff:fe78:33db) (Ping timeout: 252 seconds) |
| 2025-03-24 21:20:01 +0100 | petrichor | (~znc-user@user/petrichor) petrichor |
| 2025-03-24 21:14:54 +0100 | alp | (~alp@2001:861:8ca0:4940:92ce:4d9a:3c9b:8560) |
| 2025-03-24 21:12:42 +0100 | alfiee | (~alfiee@user/alfiee) (Ping timeout: 246 seconds) |
| 2025-03-24 21:08:41 +0100 | alfiee | (~alfiee@user/alfiee) alfiee |
| 2025-03-24 21:05:03 +0100 | notdabs | (~Owner@2600:1700:69cf:9000:c43:fe75:997:32d7) |
| 2025-03-24 20:58:07 +0100 | kh0d | (~kh0d@212.200.65.82) kh0d |
| 2025-03-24 20:57:36 +0100 | petrichor | (~znc-user@user/petrichor) (Ping timeout: 276 seconds) |
| 2025-03-24 20:56:40 +0100 | kh0d | (~kh0d@212.200.65.82) (Remote host closed the connection) |
| 2025-03-24 20:54:19 +0100 | nckx | (nckx@libera/staff/owl/nckx) (Ping timeout: 608 seconds) |
| 2025-03-24 20:46:15 +0100 | kh0d | (~kh0d@212.200.65.82) kh0d |
| 2025-03-24 20:44:23 +0100 | kh0d | (~kh0d@212.200.65.82) (Remote host closed the connection) |
| 2025-03-24 20:44:07 +0100 | <Athas> | Er, maintain an immutable array. |
| 2025-03-24 20:43:02 +0100 | <tomsmeding> | just implement it! |
| 2025-03-24 20:42:55 +0100 | <Athas> | Maybe I'm overthinking it. |
| 2025-03-24 20:42:52 +0100 | <Athas> | The algorithm is really quite trivial: https://mathworld.wolfram.com/DeterminantExpansionbyMinors.html |
| 2025-03-24 20:42:29 +0100 | <Athas> | So maybe it is good enough to maintain a mutable array and then just a bunch of linked lists. |
| 2025-03-24 20:42:14 +0100 | <Athas> | Yes. But the imperative solution is to essentially encode the matrix as a linked list of row indexes and column indexes, such that e.g. R[i] is the physical index of the successor to row 'i'. |
| 2025-03-24 20:41:11 +0100 | <tomsmeding> | so it may well be that they need to be fast after all, even if the data is small |
| 2025-03-24 20:40:59 +0100 | <tomsmeding> | because you'll probably also be doing O(n!) of those matrix manipulations |
| 2025-03-24 20:40:46 +0100 | <tomsmeding> | Athas: perhaps the dataset will be small, but even then, what really matters here is the _proportion_ of time that you spend in matrix manipulations |
| 2025-03-24 20:39:57 +0100 | machinedgod | (~machinedg@d108-173-18-100.abhsia.telus.net) (Ping timeout: 252 seconds) |
| 2025-03-24 20:39:51 +0100 | <tomsmeding> | hah, O(n!) is brutal |
| 2025-03-24 20:39:38 +0100 | <Athas> | Of course, the lovely part of an O(n!) algorithm is that the dataset will necessarily be so small that perhaps an inefficient matrix representation is unimportant. |
| 2025-03-24 20:39:15 +0100 | <Athas> | This is not a good algorithm (the complexity is something like O(n!)), but what's worse is that it depends on recursively removing rows and columns from a matrix. This can be done in efficient ways with mutable arrays, but I haven't cracked a nice way to do it in Haskell. |
| 2025-03-24 20:38:21 +0100 | <Athas> | Btw, I found that Haskell did pretty well for implementing a differentiable Runge-Kutta ODE solver, but now I'm trying to work out how to implement computation of a matrix determinant by minors. |
| 2025-03-24 20:37:41 +0100 | <Athas> | Yes, that is much better than rejecting esoterica. |
| 2025-03-24 20:36:36 +0100 | <tomsmeding> | you could even mark certain tools as "experimental", and not show them by default (or make it easy to filter them out or something) |
| 2025-03-24 20:36:23 +0100 | Sayman | (~Sayman@2401:4900:1ca3:94e0:d788:422d:b642:f1c) (Client Quit) |
| 2025-03-24 20:36:22 +0100 | <Athas> | As long as it fits in a Docker container, I would say it should go in. |