nissy-core/tools/001_derive_h48, branch python-callbackThe "engine" of nissy, including the H48 optimal solver.https://git.tronto.net/nissy-core/atom?h=python-callback2025-05-19T15:45:14ZRefactored checkdata2025-05-19T15:45:14ZSebastiano Trontosebastiano@tronto.net2025-05-19T15:45:14Zurn:sha1:62d87e063318cc4c842b1b2d8c184f48aeaf6659
Relevant changes include:
- Changed the signature of nissy_checkdata().
- Removed expected_distribution.h from tools; this data is now included
in each solver's src/ code.
- Removed distribution check for cocsep; may add back later.
Restore logging to stderr for tools2025-04-17T12:36:47ZSebastiano Trontosebastiano@tronto.net2025-04-17T12:36:47Zurn:sha1:c182f3f16e56e9533060ffa06b25391e239a28eeBetter logging function2025-04-15T13:51:40ZSebastiano Trontosebastiano@tronto.net2025-04-15T13:51:40Zurn:sha1:b25a939e2d8b045c083caf3264c4e5aac1cf88fb
Now it is possible to provide some data together with the logging function.
This is useful for example in C++, where I can now provide an arbitrary
callable object as data, and a simple wrapper function that call the
callable object as logging function.
Added buffer sizes in API args2024-10-11T20:10:42ZSebastiano Trontosebastiano@tronto.net2024-10-11T17:24:15Zurn:sha1:c4f64cf2556c0f8597e1fbc19d7c664b8da94612Removed "options" from solver selection2024-10-10T12:30:39ZSebastiano Trontosebastiano@tronto.net2024-10-10T12:30:39Zurn:sha1:f8d247c53c3c3bd94fed645b5b47a9096cadc123
Now the solver name fully determines the solver (and options)
to be used. For example, now one must specify "h48h0k4" as the
name of the solver.
This PR also fixes a couple of things in tools.
Added documentation on how to run derivedata2024-10-04T07:03:37ZSebastiano Trontosebastiano@tronto.net2024-10-04T07:03:37Zurn:sha1:fd111a1a51b8293f79db6284331af50824df92b9Clean up tool output2024-10-04T06:58:47ZSebastiano Trontosebastiano@tronto.net2024-10-04T06:58:47Zurn:sha1:32009294d2de3e197c8d4dc950559ff948ea2fc9Made table derivation tool more flexible2024-09-27T14:37:55ZSebastiano Trontosebastiano@tronto.net2024-09-27T14:37:55Zurn:sha1:41be2d294e5b6f55d485d635065096f274bc89c1
So apparently my RAM is broken. That took me a while to figure out.
While I get a replacement, I have to restrict myself to a weaker
test for the intermediate tables: instead of deriving them from
the huge table and checking that they are the same, I have to derive
a small h0k2 table from the intermediate ones and check that it is
correct. This is not a 100% proof of correctness, but it is good
enough (and much faster).