Skip to main content

Timeline for One OEIS after another

Current License: CC BY-SA 3.0

17 events
when toggle format what by license comment
Dec 11, 2017 at 13:42 comment added user202729 FYI, I have a Python program (modify from yours) that can calculate exactly terms up to 14. (that is, 6 first terms 1, 1, 1, 2, 5, 14, others take too much time for me) --- If you use Py(th)on for this answer instead, probably others can use Proton in next answers? --- However, because this is your answer, you would probably prefer to work it out yourself.
Dec 4, 2017 at 14:25 comment added hyperneutrino @NieDzejkob I'll actually probably see if I can fix Proton first before retrying this. I'll probably switch to using the builtin if possible.
Dec 4, 2017 at 14:03 comment added Maya @HyperNeutrino using a builtin function that's more than 20 times faster must help. The only problem is the generator -> list conversion bug in Proton.
Dec 4, 2017 at 13:19 comment added hyperneutrino @NieDzejkob Hm strange. I may need to just brute-force this then because I don't think optimizations will help me much anymore (or be correct for that matter :P). I'll try to fix it before most people notice :P
Dec 4, 2017 at 8:27 comment added Maya I tried rewriting this in Python, I cross-checked it like 10 times now. I couldn't find a bug, but the result for input 5 (n = 8) is 44, instead of the 14 expected. I left the Proton program running overnight, but it didn't finish. Using itertools' permutations and set union and intersection should speed it up a lot, but there seems to be a bug in the Proton interpreter. TL;DR: This answer may need corrections...
Oct 10, 2017 at 15:27 comment added Christian Sievers So you intended the graphs to be planar, and the same generation seems to work. Then everything seems fine!
Oct 10, 2017 at 15:06 comment added hyperneutrino @ChristianSievers math.stackexchange.com/a/2463430/457091
Oct 10, 2017 at 15:03 comment added Christian Sievers Maybe it doesn't matter, maybe your code does something I didn't see. Are your graphs supposed to be triangulations of the sphere? Where did you get your ideas? As for correct results, the paper also implies that (at least if you have a sphere with all the faces) you don't need the splitting of pentagons before n=12.
Oct 10, 2017 at 14:51 comment added hyperneutrino @ChristianSievers I don't think so? I mean the result for 3, 4, 5, 6 are all correct lol
Oct 10, 2017 at 14:37 comment added Christian Sievers The generation looks like what is suggested by Bowen and Fisk in "Generation of triangulations of the sphere", but since you start with a single triangle you seem to miss one face?
Oct 10, 2017 at 12:08 history edited hyperneutrino CC BY-SA 3.0
deleted 4 characters in body
Oct 10, 2017 at 12:01 comment added hyperneutrino @PeterTaylor Assuming I understand the approach you're describing, yes, it looks for triangles and places a vertex adjacent to all 3 vertices, or two adjacent cycles and deletes the common edge and places a vertex adjacent to all 4, same for 3 triangles on a pentagon. I think that's one you're describing.
Oct 10, 2017 at 6:50 comment added Peter Taylor Is this the approach of splitting triangles, squares, and pentagons as per plantri? Looks like it might be, but some of the syntax is unfamiliar.
Oct 10, 2017 at 3:21 history edited hyperneutrino CC BY-SA 3.0
added 9 characters in body
Oct 10, 2017 at 2:58 comment added hyperneutrino @Stephen Currently bugfixing lol
Oct 10, 2017 at 2:54 comment added Stephen Wait, you actually did it? If you don't write a paper with these freaking programs and go talk to some professor, you're passing up on something cool :P
Oct 10, 2017 at 2:33 history answered hyperneutrino CC BY-SA 3.0