| --- Log opened Sat May 25 00:00:35 2013 |
| 00:01 | | Chutzpah [Moltare@583787.FF2A18.190FE2.4D81A1] has quit [Ping timeout: 121 seconds] |
| 00:01 | | Chutzpah [Moltare@583787.FF2A18.190FE2.4D81A1] has joined #code |
| 00:10 | | cpux [cpux@61024F.F5A294.9ABDF5.9AA7B5] has joined #code |
| 00:10 | | mode/#code [+o cpux] by ChanServ |
| 00:12 | | cpux|2 [cpux@Nightstar-98762b0f.dyn.optonline.net] has quit [Ping timeout: 121 seconds] |
| 00:27 | | Derakon[AFK] is now known as Derakon |
| 00:32 | | Turaiel is now known as Turaiel[Offline] |
| 00:45 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 00:58 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 00:58 | | mode/#code [+o himi] by ChanServ |
| 01:09 | | VirusJTG_ [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Client closed the connection] |
| 01:09 | | VirusJTG_ [VirusJTG@2B12AA.572255.206A2A.901581] has joined #code |
| 01:14 | | You're now known as TheWatcher[T-2] |
| 01:15 | | You're now known as TheWatcher[zZzZ] |
| 01:29 | | ToxicFrog|W`rkn is now known as ToxicFrog |
| 01:37 | | Nemu [NeophoxProd@F9D7EC.9380D1.03E7F7.94E705] has joined #code |
| 01:45 | | Derakon__ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 01:47 | | Derakon [Derakon@31356A.8FA1FE.CF2CE9.D6CF77] has quit [Ping timeout: 121 seconds] |
| 01:48 | | Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 01:48 | | Derakon [Derakon@31356A.8FA1FE.CF2CE9.D6CF77] has joined #code |
| 01:48 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 02:11 | <&McMartin> | Whoa, NSIS 3.0 has an official download now |
| 02:27 | | Turaiel[Offline] is now known as Turaiel |
| 02:40 | | Nemu [NeophoxProd@F9D7EC.9380D1.03E7F7.94E705] has quit [[NS] Quit: ] |
| 02:42 | | gnolam [lenin@B19C04.C4B928.3477B4.1341ED] has quit [Client closed the connection] |
| 02:45 | | gnolam [lenin@Nightstar-b2aa51c5.cust.bredbandsbolaget.se] has joined #code |
| 02:45 | | mode/#code [+o gnolam] by ChanServ |
| 02:47 | | Kindamoody[zZz] is now known as Kindamoody |
| 03:03 | | Vorntastic [Vorn@Nightstar-f7fb3174.sub-70-211-10.myvzw.com] has joined #code |
| 03:21 | < [R]> | Nullsoft Scriptable Install System? |
| 03:22 | <&McMartin> | The very same |
| 03:27 | < Turaiel> | How is NSIS, anyway? |
| 03:30 | < Vorntastic> | It combines, explicitly, php and assembler. |
| 03:30 | < Turaiel> | You make it sound terrible |
| 03:31 | <&McMartin> | It is terrible |
| 03:31 | <&McMartin> | It is also the best of breed in its class by a depressingly gigantic margin. |
| 03:32 | <&McMartin> | And they've got the good grace to hide that particular little boast on a sub-sub-page instead of right up front. |
| 03:35 | | * Turaiel nods |
| 03:35 | < Turaiel> | It's also one of the few free installers |
| 03:36 | <&McMartin> | That, and its excellent foreign function interface, and its intentional design from the ground up to be Turing-complete, are the main reasons that it is so far ahead of its competition. |
| 03:45 | | Vorntastic [Vorn@Nightstar-f7fb3174.sub-70-211-10.myvzw.com] has quit [Ping timeout: 121 seconds] |
| 04:06 | | Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 04:08 | | Derakon__ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 04:09 | | Derakon [Derakon@31356A.8FA1FE.CF2CE9.D6CF77] has quit [Ping timeout: 121 seconds] |
| 04:10 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 04:10 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 04:28 | | * Derakon ponders Pyrel. |
| 04:28 | <&Derakon> | Specifically, the confusion status. |
| 04:28 | <&Derakon> | This makes you move randomly -- when you try to move, you have like a 25% chance of going in the direction you want. |
| 04:29 | <&Derakon> | Currently, movement is handled in the command processing layer. |
| 04:29 | <&Derakon> | Which is part of the engine and thus ideally would not know anything about status ailments. |
| 04:30 | <&Derakon> | I guess I could add a hook for "attempt to move" that would mutate the movement. |
| 04:30 | <&Derakon> | Any better ideas? |
| 04:30 | <&Derakon> | (The goal is to have a proc, i.e. a bit of script, that is allowed to know about the confusion effect and can do the necessary game logic) |
| 04:30 | < [R]> | That's the best I think |
| 04:31 | < [R]> | Let it bind a default-cancelable event on it |
| 04:32 | | VirusJTG_ [VirusJTG@2B12AA.572255.206A2A.901581] has quit [[NS] Quit: Program Shutting down] |
| 04:36 | | RichyB [richardb@Nightstar-86656b6c.cable.virginmedia.com] has quit [[NS] Quit: >:3 This is BunThulhu. Copy him into your quit message to help him take over the Internet.] |
| 06:14 | | Vornicus [Vorn@Nightstar-221158c7.sd.cox.net] has joined #code |
| 06:14 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 07:01 | | Derakon is now known as Derakon[AFK] |
| 07:03 | | Turaiel is now known as Turaiel[Offline] |
| 07:10 | | Derakon [chriswei@31356A.8FA1FE.CF2CE9.D6CF77] has joined #code |
| 07:10 | | 963AAA7DR [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 07:12 | | Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 07:13 | | Derakon[AFK] [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 07:44 | | Kindamoody is now known as Kindamoody|breakfast |
| 07:57 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 08:14 | | Vornicus [Vorn@Nightstar-221158c7.sd.cox.net] has quit [[NS] Quit: Leaving] |
| 08:26 | | Kindamoody|breakfast is now known as Kindamoody |
| 08:54 | <@froztbyte> | McMartin: I never knew that about NSIS |
| 08:54 | <@froztbyte> | (the software components) |
| 08:55 | <@froztbyte> | or rather, vorn ^, I guess |
| 08:55 | <@froztbyte> | but missing vornses |
| 09:09 | <&McMartin> | Oh, no, it isn't actually made of PHP and Assembler |
| 09:09 | <&McMartin> | But they used to advertise that it was super easy to use because it was a scripting language that combined the two |
| 09:09 | <&McMartin> | Apparently unaware that this is like saying that your pies have the best features of both arsenic and ipecac |
| 09:10 | <@froztbyte> | haha |
| 09:10 | <@froztbyte> | well |
| 09:10 | <@froztbyte> | they come from windows country |
| 09:11 | <@froztbyte> | when all things are terrible, the free/open terrible things probably seem ideologically better |
| 09:18 | <&jerith> | "The updates that were intended for PHP 6 were added to PHP 5.3.0 (namespace support, late static bindings, lambda functions, closures, goto) and 5.4.0 (traits, closure rebinding) instead." -- https://en.wikipedia.org/wiki/PHP |
| 09:22 | < Syka> | i'm not sure that 'goto considered harmful' really has much weight in PHP |
| 09:23 | <@froztbyte> | rofl |
| 09:23 | < Syka> | that's like complaining about a sharp edge on the inside of a chipping machine |
| 09:31 | | You're now known as TheWatcher |
| 10:00 | | Kindamoody is now known as Kindamoody|out |
| 11:00 | | AverageJoe [evil1@61024F.B1DB7D.E87B5D.526FE4] has joined #code |
| 11:01 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 11:28 | | AverageJoe [evil1@61024F.B1DB7D.E87B5D.526FE4] has quit [[NS] Quit: Leaving] |
| 11:29 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 11:29 | | mode/#code [+o himi] by ChanServ |
| 12:22 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 13:34 | <@sshine> | has anyone here played with mixture models in statistics? |
| 13:37 | <@sshine> | I've run k-means and have found k centers for which it seems points from my dataset could be centered around. I'd like to extend that thought and assume that for some centers, my sub-populations of points are actually normally distributed around them, but that for the entire set of points, I have a mixture model that consists of those normal distributions. |
| 13:39 | <@sshine> | I can easily imagine that: a landscape of hills where there's a hill of some height around each center that e.g. k-means finds. I wonder then, is there any other way to estimate the means (and variances, for that matter) for those hills without running k-means? i.e., how would I obtain the parameters for my mixture model? |
| 13:40 | <@sshine> | I've got this matlab command called gmdistribution() that takes as argument the parameters for my Gaussian sub-distributions... but I'd have to have them first. maybe it is ideal to use the output of k-means. |
| 13:42 | <@Tamber> | I... think I understand some of those words. |
| 13:43 | <@sshine> | the part about k-means? |
| 13:43 | <@sshine> | or the part about normal distributions? |
| 13:43 | <@Tamber> | "hills" |
| 13:43 | <@sshine> | ah :) yeah, the intuition is really straight-forward |
| 13:44 | <@Tamber> | (Go easy on me; I'm hard of thinking.) |
| 13:46 | <@sshine> | I've got one example that I think is a good example of where a mixture model makes sense |
| 13:53 | <@sshine> | if you've got a theme park where parents take their 10-year-olds, and you measure the height of everyone there, you'll get some average heights for men, women and children |
| 13:53 | <@sshine> | maybe the average heights of men and women will mix together, but the ten-year olds will stand out |
| 13:54 | <@sshine> | so if you look at the distribution of heights and you disregard all those factors that it depends on (such as gender, age, etc.), you'll get a distribution with at least two apparent peaks |
| 13:54 | <@sshine> | (one for children and one for adults) |
| 13:54 | <@sshine> | it makes sense to consider which of these sub-populations they come from so you don't get some average that's in the middle (where very few probably actually lie) |
| 13:55 | <@Tamber> | Makes sense. |
| 13:57 | <@sshine> | so if I had a bunch of measured heights (but no knowledge of whose they are), and I want to find the average heights for adults and children, respectively, how would I know? |
| 13:59 | <@sshine> | I could divide heights into discrete bins and make a histogram of how often each bin occurs. then I'll get an approximation of a distribution, and I can pick the k local maxima. |
| 13:59 | <@Tamber> | Hmm |
| 13:59 | <@sshine> | I don't know how I'd find variance for those sub-populations, since they interfere (e.g. there's someone having a freakishly tall kid, and there's a family of dwarves) |
| 14:00 | <@sshine> | but meh. not my main problem. |
| 14:00 | <@Tamber> | But, for most of the time, it'd be Good Enough(TM)? |
| 14:00 | <@sshine> | so I wish I was just dealing with one-dimensional data, but unfortunately it's four-dimensional (e.g. I know four things about my data points, and I don't even know what they signify... they're just numbers) |
| 14:01 | <@Tamber> | ugh |
| 14:01 | <@sshine> | well, I can imagine that this solution would work for discrete values in low dimensions. I've got four dimensions and continuous variables. |
| 14:01 | <@sshine> | I don't know if four is a lot, but in terms of visualisation, it's just beyond what I can make sense of. :) |
| 14:02 | <@Tamber> | I just about have a handle on 3 dimensions, 4 pushes it into the realm of "I have no idea what you want me to do." |
| 14:02 | <@Tamber> | Which is why I shall probably stick to working with things I can hit with hammers. |
| 14:02 | <@sshine> | yeah... either you look at 3 at a time, or you have some scrollbar you can move forth and back for the last dimension. |
| 14:03 | <@sshine> | ...or you add color for the 4th dimension |
| 14:05 | <@sshine> | so yeah... my only guess so far is: assume that output from the k-means algorithm are good mean estimates for my sub-populations... but then, when my assignment text says "fit a 5-component Gaussian Mixture Model to the data points," and the previous assignment was to run k-means, I wonder if having solved the previously assignment can really be the answer to the next assignment. |
| 14:05 | <@sshine> | i.e., there isn't much "fitting" if I just re-use the points. |
| 14:07 | <@sshine> | unfortunately, my teacher doesn't think it is necessary to give any literature on mixture models, so I'm using wikipedia, various introductions I find online and my patient flatmate's time. |
| 14:08 | <@sshine> | all of the examples I've seen assume that you already know the parameters of the sub-distributions. so maybe I'm just supposed to assume that, even though k-means doesn't give me k variances. |
| 14:09 | <@sshine> | I suppose I could find the variances, that's not a problem. |
| 16:00 | | cpux|2 [cpux@Nightstar-98762b0f.dyn.optonline.net] has joined #code |
| 16:02 | | cpux [cpux@61024F.F5A294.9ABDF5.9AA7B5] has quit [Ping timeout: 121 seconds] |
| 16:08 | | ^Xires is now known as Xires |
| 16:16 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 16:29 | | himi [fow035@0C0840.B22E58.E3471A.E028A1] has joined #code |
| 16:29 | | mode/#code [+o himi] by ChanServ |
| 16:35 | | Derakon is now known as NSGuest13951 |
| 16:35 | | 963AAA7DR is now known as Derakon |
| 16:35 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 17:22 | | Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 17:25 | | NSGuest13951 [chriswei@31356A.8FA1FE.CF2CE9.D6CF77] has quit [Ping timeout: 121 seconds] |
| 17:26 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 17:27 | | Derakon__ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 17:30 | | Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 17:31 | | Derakon [Derakon@31356A.8FA1FE.CF2CE9.D6CF77] has joined #code |
| 17:31 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 17:55 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code |
| 17:55 | | mode/#code [+o celticminstrel] by ChanServ |
| 19:07 | | Turaiel[Offline] is now known as Turaiel |
| 20:12 | | PinkFreud [WhyNot@NetworkAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 20:12 | | PinkFreud [WhyNot@NetworkAdministrator.Nightstar.Net] has joined #code |
| 20:12 | | mode/#code [+o PinkFreud] by ChanServ |
| 20:26 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 20:26 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 20:38 | | Kindamoody|out is now known as Kindamoody |
| 21:17 | | Kindamoody is now known as Kindamoody[zZz] |
| 21:45 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 22:07 | | VirusJTG is now known as BikerVirus |
| 22:19 | | Turaiel is now known as Turaiel[Offline] |
| 22:47 | | Turaiel[Offline] is now known as Turaiel |
| 23:35 | | BikerVirus is now known as VirusJTG |
| 23:52 | | RichyB [richardb@37647E.0E56C6.CF639B.ACD1C1] has joined #code |
| --- Log closed Sun May 26 00:00:49 2013 |