--- Log opened Thu Apr 28 00:00:32 2016 | ||
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 252 seconds] | 00:24 | |
-!- WiggleRoom_ [49bd0ad0@gateway/web/freenode/ip.73.189.10.208] has joined #shogun | 01:40 | |
WiggleRoom_ | Hey I have a machine learning question! | 01:41 |
---|---|---|
-!- WiggleRoom__ [49bd0ad0@gateway/web/freenode/ip.73.189.10.208] has joined #shogun | 01:42 | |
WiggleRoom__ | Is anyone there | 01:42 |
WiggleRoom__ | My model is huuuuge like 100GB AND GROWING! How can I store and use it? It won't even load into memory! | 01:42 |
-!- WiggleRoom_ [49bd0ad0@gateway/web/freenode/ip.73.189.10.208] has quit [Ping timeout: 250 seconds] | 01:46 | |
-!- WiggleRoom__ [49bd0ad0@gateway/web/freenode/ip.73.189.10.208] has quit [Quit: Page closed] | 01:50 | |
-!- lupinix [~quassel@fedora/lupinix] has quit [Ping timeout: 240 seconds] | 01:50 | |
-!- lupinix [~quassel@v22014041761818086.yourvserver.net] has joined #shogun | 01:51 | |
-!- lupinix [~quassel@v22014041761818086.yourvserver.net] has quit [Changing host] | 01:51 | |
-!- lupinix [~quassel@fedora/lupinix] has joined #shogun | 01:51 | |
-!- mizari [~mizari@95-174-213-100.nts.su] has joined #shogun | 05:13 | |
-!- arianepaola [~ariane@unaffiliated/arianepaola] has quit [Excess Flood] | 06:21 | |
-!- arianepaola [~ariane@unaffiliated/arianepaola] has joined #shogun | 06:22 | |
-!- Saurabh7 [~Saurabh7@103.225.100.51] has quit [Quit: Leaving] | 06:38 | |
-!- besser82 [~besser82@fedora/besser82] has joined #shogun | 08:03 | |
-!- mode/#shogun [+o besser82] by ChanServ | 08:03 | |
@wiking | lisitsyn: pingpong | 09:34 |
sonne|work | WiggleRoom__ wrong timezone :P | 09:38 |
@wiking | :) | 09:39 |
@wiking | sonne|work: did you guys ever wondered to support mixed types in Features? :) | 09:39 |
sonne|work | wiking: shogun does that pretty well already... | 09:40 |
sonne|work | and I use this quite often | 09:40 |
@wiking | where? | 09:40 |
@wiking | which Features? | 09:40 |
sonne|work | CombinedFeatures | 09:40 |
sonne|work | strings | 09:40 |
sonne|work | realfeatures | 09:40 |
@wiking | doh | 09:40 |
sonne|work | etc | 09:40 |
@wiking | so you put combinedFeatures everything | 09:41 |
@wiking | ? | 09:41 |
@wiking | i've used this only with combinedKernel | 09:41 |
sonne|work | we even have combined dotfeatures | 09:41 |
sonne|work | so you can train e.g. a linear svm with all kinds of DotFeatures attached to it | 09:41 |
sonne|work | sure | 09:41 |
@wiking | never thought of actually | 09:41 |
sonne|work | some methods required RealFeatures though | 09:42 |
@wiking | using that as combination of string and real fatures | 09:42 |
@wiking | yeah | 09:42 |
sonne|work | that is one of the strengths of shogun | 09:42 |
@wiking | that's ok | 09:42 |
sonne|work | but it is not just strings | 09:42 |
sonne|work | it is boolean/int/float/double ... you can mix that at will | 09:42 |
@wiking | yeah yeah | 09:42 |
@wiking | but yeah we should ahve an example for this | 09:43 |
@wiking | because i tooootally had no clue | 09:43 |
@wiking | that actually the combinedfeatures | 09:43 |
@wiking | could be used like this | 09:43 |
@wiking | maybe that's only me | 09:43 |
@wiking | but then again | 09:43 |
@wiking | mmm | 09:44 |
@wiking | the only thing would be good to have i guess | 09:44 |
@wiking | and IO method | 09:44 |
@wiking | that actually can create a CombinedFeatures | 09:44 |
@wiking | from a csv | 09:44 |
@wiking | detecting the types | 09:44 |
@wiking | right? | 09:44 |
@wiking | since basically we can say that CombinedFeatures = DataFrame | 09:44 |
@wiking | but now you would have to manually do the loading of a csv with various types of features | 09:45 |
@wiking | into a CombinedFeatures | 09:45 |
@wiking | OXPHOS could work on this | 09:46 |
@wiking | dear god, all aws emails are ending up on gmail in spam | 09:47 |
@wiking | g really does not want people to use aws :) | 09:47 |
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 276 seconds] | 10:16 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Remote host closed the connection] | 10:17 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 10:36 | |
-!- mode/#shogun [+o wiking] by ChanServ | 10:37 | |
@wiking | haha | 11:06 |
@wiking | just realised our welcome string for #shogun: "[#shogun] Welcome to #shogun - The official channel of the shogun machine learning toolbox - http://www.shogun-toolbox.org/. We participate in GSoC 2014 http://bit.ly/16LaK5Z. This channel is logged and logs are available from http://bit.ly/eQDk8Q. | 11:07 |
@wiking | note: "We participate in GSoC 2014" | 11:07 |
@wiking | :) | 11:07 |
lisitsyn | lol | 11:07 |
lisitsyn | wiking: change it, you're an op | 11:07 |
lisitsyn | I've lost my registration actually | 11:07 |
@wiking | where? :) | 11:08 |
@wiking | i mean where did you lost it | 11:08 |
@wiking | ? | 11:08 |
lisitsyn | nickserv | 11:08 |
@wiking | ok done | 11:09 |
@wiking | if you check the link of We participate in GSoC 2014 | 11:09 |
@wiking | it is actually pointing to a 2013 page | 11:09 |
@wiking | :) | 11:09 |
@wiking | so it's really good : | 11:10 |
@wiking | :> | 11:10 |
lisitsyn | 2013! | 11:10 |
lisitsyn | рфрф | 11:10 |
lisitsyn | haha | 11:10 |
lisitsyn | oops | 11:10 |
@wiking | anyhow now it's updated | 11:10 |
lisitsyn | it is hard to live with multiple keyboard layouts :D | 11:10 |
@wiking | darn cyrillic :) | 11:10 |
lisitsyn | wiking: there was an initiative | 11:11 |
lisitsyn | to use latin letters | 11:11 |
@wiking | it's easy | 11:11 |
@wiking | there's a 1-1 mapping | 11:11 |
lisitsyn | exactly | 11:11 |
@wiking | (at least we have in serbian) | 11:11 |
@wiking | when i told this to an american | 11:11 |
lisitsyn | yeah there are hard cases though | 11:11 |
@wiking | he tought i'm crazy | 11:11 |
@wiking | :) | 11:11 |
@wiking | but then again | 11:11 |
@wiking | :))))) | 11:11 |
lisitsyn | iirc you can get some ambiguity with sh sch | 11:12 |
@wiking | living with 26 characters in your alphabet is quite weak :P | 11:12 |
@wiking | we have 44 in hugnarian | 11:13 |
@wiking | 33 in russian right? | 11:13 |
lisitsyn | yes | 11:13 |
lisitsyn | 33 | 11:13 |
lisitsyn | most of them are not that useful though | 11:13 |
lisitsyn | ь/ъ | 11:14 |
@wiking | so yeah it's easy :) | 11:14 |
lisitsyn | legacy stuff | 11:14 |
lisitsyn | е/ё | 11:14 |
@wiking | if you have a normal alphabet to cover all of them | 11:14 |
@wiking | :P | 11:14 |
lisitsyn | wiking: american alphabet must be of like ten letters actually | 11:14 |
lisitsyn | they like to simplify things :P | 11:14 |
@wiking | :) | 11:15 |
@wiking | yeah all the vowels are out | 11:15 |
@wiking | :> | 11:15 |
@wiking | flckr | 11:15 |
@wiking | rzr | 11:15 |
@wiking | :) | 11:15 |
lisitsyn | twttr | 11:15 |
@wiking | yeap pretty much | 11:15 |
lisitsyn | ggl | 11:15 |
-!- HeikoS [~heiko@p57972783.dip0.t-ipconnect.de] has joined #shogun | 11:15 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 11:15 | |
lisitsyn | hks | 11:15 |
@wiking | :DDDD | 11:16 |
lisitsyn | h hks | 11:16 |
lisitsyn | hk! | 11:16 |
lisitsyn | hll | 11:16 |
lisitsyn | H! | 11:16 |
lisitsyn | v lst m vwls | 11:16 |
lisitsyn | dmn | 11:17 |
@wiking | wll y knw n hbrw thrs n vwls jst dts | 11:17 |
@wiking | :) | 11:17 |
lisitsyn | xctl | 11:17 |
lisitsyn | HeikoS: H! | 11:18 |
-!- lisitsyn is now known as lstsn | 11:19 | |
lstsn | HkS | 11:19 |
-!- HeikoS [~heiko@p57972783.dip0.t-ipconnect.de] has quit [Ping timeout: 260 seconds] | 11:21 | |
lstsn | dmn | 11:22 |
-!- HeikoS [~heiko@p57972783.dip0.t-ipconnect.de] has joined #shogun | 11:26 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 11:26 | |
@wiking | :> | 11:27 |
@HeikoS | wiking: have an appointment to sign things next Tuesday | 11:30 |
@HeikoS | wiking: they will check all the docs now, I brought them in yesterday | 11:30 |
@HeikoS | sonney sent me the original signatures per post-mail as they are needed | 11:30 |
@HeikoS | So looking good | 11:30 |
lstsn | cl | 11:40 |
lstsn | HkS | 11:45 |
@HeikoS | lstsn: ? | 12:02 |
lstsn | HeikoS: v lst m vwls | 12:38 |
CaBa | hi | 13:02 |
CaBa | can one adjust the number of threads used by libsvm somehow? | 13:03 |
@HeikoS | CaBa: I dont think so | 13:03 |
@HeikoS | CaBa: if you precompute the kernel matrix, then multiple threads are used | 13:04 |
@HeikoS | lstsn: ! | 13:04 |
lstsn | HeikoS: wazzzzup | 13:04 |
@HeikoS | lstsn: !!! | 13:04 |
lstsn | wzzzzzuppp | 13:04 |
CaBa | HeikoS: i'm wondering about finetuning "multiple" ;) it seems to use the number of physical CPUs rather than logical CPUs to decide how many threads to create... I wonder if hyperthreading would boost this further... | 13:05 |
@HeikoS | CaBa: Shogun always uses the number of cores available to the OS | 13:06 |
@HeikoS | hyperthreading just gives you 4 cores from 2 physical cores | 13:06 |
@HeikoS | but the OS always "sees" 4 | 13:06 |
CaBa | HeikoS: hm okay then HT simply doesn't kick in that well in case of kernel training.. | 13:40 |
@HeikoS | CaBa: I think LibSVM is simply not parallelised code | 13:40 |
@HeikoS | Only if you pre-compute the kernel matrix, this is parallelised | 13:41 |
@HeikoS | This gets important if you cross-validate | 13:41 |
CaBa | HeikoS: well i have a 12-core machine with HT and during train() the process uses ~1200% cpu ressources... so train() seems parallelized... | 13:42 |
@HeikoS | oh ok, I see | 13:42 |
@HeikoS | maybe someone added that :D | 13:42 |
@HeikoS | try sg_parallel | 13:42 |
CaBa | ;) | 13:42 |
@HeikoS | you can change the number of cores used within there | 13:43 |
@HeikoS | any_sg_object_instance->get_global_parallel()->set_num_threads(n) | 13:44 |
CaBa | HeikoS: it really seems that HT doesn't scale well here. Calling that function works, I can e.g. limit to 4 threads or so, but anything above 12 won't increase the load much | 13:49 |
@HeikoS | That makes sense I think | 13:49 |
@HeikoS | solving SVM is a sequential problem | 13:49 |
@HeikoS | LibSVM is not designed formultiple cores | 13:50 |
@HeikoS | I think there are SVM solvers that can exploit more | 13:50 |
CaBa | HeikoS: What are the criteria for choosing the backend anyway? Should I be using another more efficient packend? | 13:50 |
CaBa | *backend | 13:50 |
@HeikoS | LibSVM is a standard library | 13:50 |
@HeikoS | not the fastest | 13:50 |
@HeikoS | it depends on your problem | 13:50 |
@HeikoS | also, if you need to tune parameters, this is where most of the cpu cycles go (x-validation, grid search, etc) | 13:50 |
@HeikoS | so if you parallelise that, you get almost linear gains | 13:50 |
@HeikoS | in the number of threads, as completely independent | 13:51 |
CaBa | indeed, maybe i shouldn't be optimizing here... right now i'm not doing any model selection, just playing with a single model | 13:51 |
CaBa | but since that won't stay that way i can just rely on the model selection parallelization later... | 13:52 |
CaBa | HeikoS: is there like an overview of the different backends and how to choose it? like how does it depend on my problem? | 13:53 |
@HeikoS | I am sure there is | 13:53 |
@HeikoS | not so much in Shoguns doc though | 13:54 |
@HeikoS | Depends a bit on your problem | 13:54 |
@HeikoS | how much N, D, non-linearity you have? | 13:54 |
CaBa | right now, i'm working with ~80 numerical (real) features on ~54k observations... i might however also want to use or adopt some of the string kernels in shogun soon | 13:54 |
CaBa | right now i just played with a simple, plain linear kernel. however, nobody tells me that this data is linearly separable, hence i'll also try other kernels like poly or rbf... | 13:56 |
CaBa | i take it there is no general doitall workhorse library backend? i really didn't "choose" libsvm, i just saw it in one of the examples and plugged it in ;) | 13:57 |
@HeikoS | I see | 14:03 |
@HeikoS | so RBF kernel seems like a good idea | 14:03 |
@HeikoS | however, if you go for a linear kernel, then LibLinear ist *much* faster | 14:03 |
@HeikoS | there is a multi-core version of it available, but we haven't integrated it yet unfortunately | 14:03 |
@HeikoS | sonney2k_, sonne|work might have comments | 14:04 |
CaBa | Hm. LibLinear... doesn't show up in the SVM class inheritance diagram ^^ | 14:04 |
@HeikoS | http://shogun-toolbox.org/doc/en/latest/classshogun_1_1CLibLinear.html | 14:18 |
@HeikoS | http://shogun-toolbox.org/doc/en/latest/classshogun_1_1CLinearMachine.html | 14:18 |
@HeikoS | SVMLight should also be faster than libsvm | 14:19 |
@HeikoS | and this guy is also fast I think: SVMOCAS | 14:19 |
CaBa | I'll play around and also check how easily interchangable they are, thanks :) | 14:21 |
sonne|work | CaBa: well try with a small subsample of your data and the rbf kernel first then once you know which parameter range (kernel width, C) seem to do well use a bigger set | 14:41 |
@wiking | mooo | 14:53 |
sonne|work | wiking: holy cow? | 15:06 |
sonne|work | wiking: apt-get moo ? | 15:06 |
CaBa | sonne|work: will do :) | 15:17 |
CaBa | (subsample that is... not that much into mooing, though i totally appreciate super cow powers...) | 15:19 |
@wiking | yes | 15:28 |
@wiking | sonne|work: hoooooooly holly cow | 15:28 |
@wiking | so what do you say about an IO func for CombinedFeatures? | 15:29 |
sonne|work | wiking: sure do it... but no idea how | 15:30 |
CaBa | replacing LibSVM by SVMLight in my python script gets me segfaults \o/ :) | 16:10 |
-!- besser82 [~besser82@fedora/besser82] has joined #shogun | 16:17 | |
-!- mode/#shogun [+o besser82] by ChanServ | 16:17 | |
@wiking | :) | 16:51 |
-!- OXPHOS [8ca3fe9e@gateway/web/freenode/ip.140.163.254.158] has joined #shogun | 16:53 | |
OXPHOS | Hey @ Heiko, wiking, I'll be analyzing data basically the whole day. We can discuss about the project anytime as long as my PI doesn't jump out. | 17:02 |
-!- besser82 [~besser82@fedora/besser82] has quit [Ping timeout: 246 seconds] | 17:08 | |
@HeikoS | OXPHOS: jojo | 17:45 |
OXPHOS | HeikoS: hello | 17:46 |
@HeikoS | OXPHOS: so I think first step is to share a googledoc with your proposal | 17:47 |
@HeikoS | it should have two parts | 17:47 |
@HeikoS | 1. schedule of things we want to do and how long they take. This is "what and when" | 17:47 |
@HeikoS | 2. A detailed list of issues that we have to solve. Dividing tasks into subtasks, figuting out details etc. This is "how exactly things are done" | 17:47 |
@HeikoS | could you start a doc (doesnt have to be perfect for now, proposal is ok) and share it? | 17:48 |
OXPHOS | Sure. Thought the current one is already shared doc. Be back in 1 sec | 17:49 |
OXPHOS | It has to be .word right? | 17:49 |
OXPHOS | .doc | 17:49 |
@HeikoS | maybe we should create a new doc | 17:49 |
@HeikoS | and paste your proposal in, we dont want to change the proposal doc | 17:49 |
@HeikoS | googledoc .doc yes | 17:49 |
OXPHOS | Yep make sense | 17:49 |
@HeikoS | OXPHOS: cool | 17:50 |
@HeikoS | OXPHOS: one thing we will definitely need to do is to prototype a serialization using ceral. This will be done best in a separate environment, i.e. not in the shogun source tree but in a standalone file to isolate things as good as possible | 17:51 |
@HeikoS | one other thing to to discuss a few linalg internal changes with wiking and lambday | 17:51 |
OXPHOS | Copy that | 17:52 |
OXPHOS | I sent you the docs link @ HeikoS | 17:53 |
@HeikoS | yep saw it | 17:53 |
@HeikoS | I suggest we keep going slowly but steadily | 17:53 |
@HeikoS | dont need to make a lot of progress every day, but we should make constant progress, even if slow and we are busy | 17:54 |
OXPHOS | Agree | 17:54 |
OXPHOS | And also the readme files? XD | 17:54 |
@HeikoS | readme files? | 17:55 |
OXPHOS | There's a readme.linalg, with several 'coming soon' | 17:55 |
@HeikoS | ah yes | 17:56 |
@HeikoS | definitely should look at those | 17:56 |
@HeikoS | after GSOC, it would be good to have some developer docs that are complete on this topic :) | 17:56 |
OXPHOS | Yes I guess things are going to change a lot | 17:56 |
-!- mizari [~mizari@95-174-213-100.nts.su] has quit [Quit: Leaving] | 17:57 | |
@HeikoS | definitely | 17:57 |
@HeikoS | oh and the final big thing is the smart pointers | 17:57 |
@HeikoS | this also needs some prototype that applies the principle to an isolated part of shogun, i.e. one single class | 17:58 |
@HeikoS | re io, we need to talk to wiking | 17:58 |
@HeikoS | need to use proper modern c++ for this, the code is horribly old ;) | 17:58 |
OXPHOS | wiking once said he's almost done with the I/O | 17:58 |
OXPHOS | Stupid question - Is it possible I use the same class/environment/whatever for serialization/smart pointer test? | 17:59 |
@HeikoS | test in what sense? | 18:00 |
OXPHOS | I mean set up the prototype | 18:03 |
@HeikoS | ah i see | 18:04 |
@HeikoS | I think it is necessary in fact | 18:04 |
@HeikoS | cereal is based on smart pointers | 18:04 |
@HeikoS | cannot deal with raw pointers | 18:04 |
@HeikoS | so in fact these two depend on each other | 18:04 |
OXPHOS | Wow. I like that | 18:05 |
OXPHOS | for no reason | 18:05 |
@HeikoS | hehe | 18:06 |
@HeikoS | that is good | 18:06 |
@HeikoS | makes so well suited for that | 18:06 |
@HeikoS | since it will be a mess :D | 18:06 |
@HeikoS | but it IS quite cool 8-) | 18:07 |
@HeikoS | getting rid of all this 90s code ;) | 18:07 |
OXPHOS | EVOLUTION! | 18:08 |
@HeikoS | exactly | 18:08 |
@HeikoS | but some shogun classes will go extinct hehe | 18:08 |
OXPHOS | like human in some day. meh. | 18:09 |
@HeikoS | better not make shogun too powerful then | 18:09 |
@HeikoS | to buy time | 18:09 |
OXPHOS | haha | 18:10 |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun | 18:50 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Ping timeout: 250 seconds] | 18:55 | |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has joined #shogun | 18:57 | |
sanuj | hello everyone! | 19:04 |
sanuj | HeikoS: are you making shogun accept donations in germany? | 19:04 |
sanuj | just curious :) | 19:04 |
@wiking | :> | 19:11 |
-!- besser82 [~besser82@fedora/besser82] has joined #shogun | 19:14 | |
-!- mode/#shogun [+o besser82] by ChanServ | 19:14 | |
@wiking | OXPHOS: sorry was away | 19:15 |
sanuj | lstsn: what happened to your name? :P | 19:41 |
CaBa | the latest release is 4.1.0, right? But there is no online doc? | 19:57 |
@wiking | he removed all the unnecessary vowels :) | 19:57 |
@wiking | CaBa: there was a bug in that release | 19:58 |
@wiking | you can use the docs of 4.2.0 (development) | 19:58 |
@wiking | it shouldn't be too different | 19:58 |
CaBa | wiking: i just started to play with model selection and the 4.2 doc says that the GaussianKernel has a "log_width" param, which it doesn't in 4.1.0 ;-) | 20:00 |
lstsn | sanuj: I've lost vowels | 20:05 |
sanuj | ohkay | 20:05 |
@wiking | CaBa: why dont u use develop branch from the git repo? | 20:05 |
CaBa | wiking: i prefer letting only homebrew touch my /usr/local and installing it anywhere else would require me to fiddle with DYLD_LIBRARY_PATH etc... but yes, i'm tempted to switch to the latest and greatest shogun at some point, maybe i'll look into altering the homebrew formula to checkout from upstream git instead of fetching a stable tarball... | 20:11 |
CaBa | wiking: you don't have nightly tarballs or such thing, right? | 20:11 |
@wiking | :> | 20:13 |
CaBa | maybe i can make use of githubs tarballs... | 20:13 |
@wiking | you can use github | 20:13 |
@wiking | and you can install it to any prefix | 20:13 |
@wiking | btw can you tell what's the use case? :) | 20:14 |
@wiking | because actually we dont know what shogun is being used for :) | 20:14 |
@wiking | :P | 20:14 |
CaBa | yeah i know it's a standard cmake or autoconf project, i remember compiling it from source a while ago. it's just trouble to use shared libaries that aren't in the standard locations the system uses to look for them... | 20:16 |
@wiking | yeah indeed | 20:17 |
-!- sanuj [0e8bc402@gateway/web/freenode/ip.14.139.196.2] has quit [Quit: Page closed] | 20:17 | |
CaBa | sure :) i'm a bioinformatician, my features are usually DNA sequences (originally) and my field of research is immunogenetics, not far from what gunnar and sören have applied ML for already | 20:18 |
CaBa | what's your background? | 20:19 |
@wiking | ah i see | 20:19 |
@wiking | i did computer vision | 20:19 |
@wiking | unfortunately not for a while now | 20:20 |
CaBa | and now? | 20:20 |
@wiking | went to industry | 20:21 |
CaBa | still involved in ML or are you spending your free time with shogun? | 20:22 |
@wiking | free time shogun | 20:22 |
@wiking | but yeah more or less still involved with ml | 20:22 |
@wiking | although nothing with shogun directly | 20:23 |
CaBa | where are you from? | 20:23 |
OXPHOS | wiking: me too. Just came back from lunch : ) | 20:29 |
@wiking | CaBa: huh that's a complicated answer | 20:29 |
@wiking | OXPHOS: cool... yeah we should do a quick demo with cereal :) | 20:30 |
OXPHOS | wiking: Sure. Lemme read more about it. | 20:32 |
CaBa | wiking: :D sounds interesting | 20:33 |
@wiking | http://motherfuckingwebsite.com/ | 20:57 |
CaBa | ^^ | 21:17 |
-!- HeikoS1 [~heiko@p4FED52C3.dip0.t-ipconnect.de] has joined #shogun | 21:35 | |
-!- HeikoS [~heiko@p57972783.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds] | 21:36 | |
-!- OXPHOS [8ca3fe9e@gateway/web/freenode/ip.140.163.254.158] has quit [Quit: Page closed] | 22:56 | |
-!- HeikoS1 [~heiko@p4FED52C3.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds] | 23:39 | |
--- Log closed Fri Apr 29 00:00:34 2016 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!