Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 231 232 233 234 235 236 237 238 239 240 241 .. 5022
— 502,152 events total
2020-09-26 17:57:51 × poljar quits (~poljar@93-139-70-179.adsl.net.t-com.hr) (Remote host closed the connection)
2020-09-26 17:58:10 <nineonine> it works fine up to 20
2020-09-26 17:58:18 <nineonine> after that it starts returning nonsense results
2020-09-26 17:58:47 → poljar joins (~poljar@93-139-70-179.adsl.net.t-com.hr)
2020-09-26 18:00:02 × rektide1 quits (~rektide@84.39.117.57) ()
2020-09-26 18:00:24 <monochrom> Could you post the complete code that produced -17.46... ?
2020-09-26 18:00:27 <monochrom> @where paste
2020-09-26 18:00:27 <lambdabot> Help us help you: please paste full code, input and/or output at eg https://paste.tomsmeding.com
2020-09-26 18:00:57 <nineonine> here you go!
2020-09-26 18:00:57 <nineonine> https://paste.tomsmeding.com/SDUUcLFg
2020-09-26 18:01:00 <nineonine> thx!
2020-09-26 18:01:29 × aplainzetakind quits (~johndoe@captainludd.powered.by.lunarbnc.net) (Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net)
2020-09-26 18:01:30 → laxask joins (~lax@unaffiliated/laxask)
2020-09-26 18:01:38 laxask is now known as sudden
2020-09-26 18:01:43 <tomsmeding> nineonine: the inner binomCoef takes Int's
2020-09-26 18:01:49 <tomsmeding> that product [1..m] is going to overflow
2020-09-26 18:01:54 <nineonine> aha
2020-09-26 18:01:58 × mu_ quits (~mu@unaffiliated/mu) (Quit: mu_)
2020-09-26 18:02:11 <nineonine> how do I fix it
2020-09-26 18:02:14 <monochrom> > product [1..21] :: Int
2020-09-26 18:02:16 <lambdabot> -4249290049419214848
2020-09-26 18:02:20 <tomsmeding> possible solutions: Integer instead of Int; Double instead of Int; use a different algorithm that isn't as susceptible to overflow :)
2020-09-26 18:02:21 <monochrom> Ah, like that.
2020-09-26 18:02:26 <fluturel> ski: are you still here?
2020-09-26 18:02:38 <ski> no
2020-09-26 18:02:46 <ski> i'm over there
2020-09-26 18:02:46 × ap5 quits (~ap5@85.184.161.18) (Ping timeout: 256 seconds)
2020-09-26 18:02:47 <monochrom> Yeah using factorial for binom is the worst idea.
2020-09-26 18:02:53 <fluturel> i think i got it with pattern matching
2020-09-26 18:02:56 <monochrom> In multiple ways, not just this one.
2020-09-26 18:03:00 <ski> fluturel : okay ?
2020-09-26 18:03:01 <nineonine> what would be a better solution here?
2020-09-26 18:03:20 <tomsmeding> nineonine: e.g. https://math.stackexchange.com/a/927064/68044
2020-09-26 18:03:22 <fluturel> ski : you know, the zip' function
2020-09-26 18:03:46 <tomsmeding> wait no
2020-09-26 18:03:55 <ski> fluturel : yes, i didn't forget :)
2020-09-26 18:04:24 <ski> (i meant "okay ?", as in "well, let's hear about it ?")
2020-09-26 18:04:33 → adam_wespiser joins (~adam_wesp@209.6.42.110)
2020-09-26 18:04:36 <monochrom> > product [1..21] :: Double
2020-09-26 18:04:38 <fluturel> zip' :: [a] -> [b] -> [(a,b)]
2020-09-26 18:04:38 <lambdabot> 5.109094217170944e19
2020-09-26 18:04:39 <fluturel> zip' (x:xs) (y:ys) = [(x,y)] ++ (zip' xs ys)
2020-09-26 18:04:39 <fluturel> zip' _ [] = []
2020-09-26 18:04:39 <fluturel> zip' [] _ = []
2020-09-26 18:04:53 <fluturel> How do i post multiline code
2020-09-26 18:05:14 <geekosaur> use a pastebin, ideally
2020-09-26 18:05:27 <geekosaur> @where paste
2020-09-26 18:05:28 <lambdabot> Help us help you: please paste full code, input and/or output at eg https://paste.tomsmeding.com
2020-09-26 18:05:29 → aplainzetakind joins (~johndoe@captainludd.powered.by.lunarbnc.net)
2020-09-26 18:05:55 <dsal> fluturel: You could also use ;
2020-09-26 18:06:09 × rcdilorenzo quits (~rcdiloren@cpe-76-182-87-188.nc.res.rr.com) (Remote host closed the connection)
2020-09-26 18:06:13 <fluturel> ski: https://pastebin.com/JLshNdsN
2020-09-26 18:06:24 <fluturel> dsal: already made a pastebin, but thank you
2020-09-26 18:06:53 <dsal> ; is helpful if you want lambdabot to run the code. :)
2020-09-26 18:07:06 → snakemasterflex joins (~snakemast@213.100.206.23)
2020-09-26 18:07:38 <ski> fluturel : yep, that looks fine :)
2020-09-26 18:07:50 <tomsmeding> nineonine: https://codeforces.com/blog/entry/55311 that's the algorithm that I know, I can try functionalising it
2020-09-26 18:07:50 <fluturel> ski: i vwas wondering if there is a better way to rewrite the last two lines. I mean,. they are the same thing, just in onecase the first list finishes its elements first
2020-09-26 18:08:13 → chaosmasttter joins (~chaosmast@p200300c4a7105f0120adba39e14d4586.dip0.t-ipconnect.de)
2020-09-26 18:08:16 <ski> fluturel : you could replace them with zip' _ _ = []
2020-09-26 18:08:19 <fluturel> or the other way around actually
2020-09-26 18:08:29 <ski> as long as that defining equation is after the recursive one
2020-09-26 18:08:33 <nineonine> tomsmeding, that would be great :)
2020-09-26 18:09:04 × HarveyPwca quits (~HarveyPwc@c-98-220-98-201.hsd1.il.comcast.net) (Quit: Leaving)
2020-09-26 18:09:12 <ski> fluturel : btw, the brackets around the recursive call are redundant. and you could use `:' in place of `++'
2020-09-26 18:09:35 <fluturel> ski: actually yeah, you are right. If it doesn't match with the first pattern, the second will catch it regardless and return an empty list
2020-09-26 18:09:44 <fluturel> ski: oh, didn't know that
2020-09-26 18:10:11 → justanotheruser joins (~justanoth@unaffiliated/justanotheruser)
2020-09-26 18:10:47 <fluturel> ski: just to clarify, which brackets? These: [(x,y)] ?
2020-09-26 18:11:25 × snakemasterflex quits (~snakemast@213.100.206.23) (Ping timeout: 246 seconds)
2020-09-26 18:11:27 <ski> fluturel : one argument for not replacing them like that would be that, the way you've written the defining equations, you could actually have them in any order. and you can understand them in isolation, as true statements about your function. but if you have the patterns in the base case overlap with the patterns in the recursive case, then you can't understand the base case in isolation
2020-09-26 18:11:36 <ski> fluturel : no, the round brackets
2020-09-26 18:11:39 <dsal> fluturel: (zip' xs ys)
2020-09-26 18:11:53 <ski> ("around the recursive call")
2020-09-26 18:11:55 <dsal> hlint will tell you things like that.
2020-09-26 18:12:26 <fluturel> is hlint for emacs a thing?
2020-09-26 18:12:32 <dsal> Just a cli tool
2020-09-26 18:12:42 <dsal> But probably.
2020-09-26 18:13:40 → acidjnk_new joins (~acidjnk@p200300d0c723789624d205acd1b15fd8.dip0.t-ipconnect.de)
2020-09-26 18:13:58 → Saten-san joins (~Saten-san@ip-213-49-124-100.dsl.scarlet.be)
2020-09-26 18:14:04 <ski> fluturel : .. but in some situations, one can replace a lot of extra cases, with just a single "catch-all"/"default" case, at the price of having overlapping cases
2020-09-26 18:14:15 <fluturel> I am going to look more into defining the !!? operator and working with Maybe, as i have something to learn from there, but i think this is the more elegant solution. Not gonna lie, i feel quite proud of it :)
2020-09-26 18:14:45 <nineonine> tomsmeding: btw, I used the solution from that SO answer and it seems to work fine. Many thanks again!
2020-09-26 18:14:46 <ski> yes, this is the idiomatic solution
2020-09-26 18:14:56 <nineonine> not SO, but SE
2020-09-26 18:15:20 <ski> fluturel : i was just thinking that, doing it without explicit pattern-matching, could possibly be an interesting exercise
2020-09-26 18:15:23 <fluturel> ski: i see what you mean, i guess for more complicated patterns, understanding them in isolation is better, but here overlapping them won't be such a burden for a reader
2020-09-26 18:15:39 <tomsmeding> nineonine: yes it does seem to work fine indeed; I'm kind of surprised that it does though
2020-09-26 18:15:40 × adam_wespiser quits (~adam_wesp@209.6.42.110) (Ping timeout: 272 seconds)
2020-09-26 18:15:47 <tomsmeding> also, it contains non-tail recursion, while this one doesn't: https://paste.tomsmeding.com/cj6pnnXi
2020-09-26 18:15:51 <ski> (btw, of course `!!' itself is defined, using pattern-matching. and the function you'd call from `!!?' also would be)
2020-09-26 18:15:53 <tomsmeding> but feel free to use the SE one :p
2020-09-26 18:16:00 <nineonine> will do, thx :D
2020-09-26 18:16:13 <tomsmeding> (oh not choose2 but binomCoef of course)
2020-09-26 18:16:27 <nineonine> I don't quite understand though why isn't it included in base
2020-09-26 18:16:35 <nineonine> it seems somewhat useful sometimes
2020-09-26 18:16:52 <nineonine> maybe because it is extremely inefficient? dunno
2020-09-26 18:17:20 <dsal> base is too big. Things in base should be very useful all the time.
2020-09-26 18:17:44 <nineonine> I agree
2020-09-26 18:17:55 <tomsmeding> there is also not really a canonical implementation
←Prev  Next→
Page 1 .. 231 232 233 234 235 236 237 238 239 240 241 .. 5022

All times are in UTC.