--- Log opened Mon Mar 19 00:00:19 2012 | ||
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 00:11 | |
-!- in3xes_ [~in3xes@180.149.49.230] has joined #shogun | 01:28 | |
-!- in3xes [in3xes@49.248.244.12] has quit [Ping timeout: 250 seconds] | 01:32 | |
-!- vikram360 [~vikram360@117.192.169.153] has joined #shogun | 01:39 | |
-!- vikram360 [~vikram360@117.192.169.153] has quit [Ping timeout: 252 seconds] | 01:46 | |
-!- vikram360 [~vikram360@117.192.172.40] has joined #shogun | 01:47 | |
-!- vikram360 [~vikram360@117.192.172.40] has quit [Ping timeout: 248 seconds] | 02:04 | |
-!- in3xes_ is now known as in3xes | 02:27 | |
-!- in3xes [~in3xes@180.149.49.230] has quit [Quit: Leaving] | 02:53 | |
-!- l0nr4n [~l0nr4n@unaffiliated/l0nr4n] has quit [Quit: zzzz...] | 03:34 | |
-!- gsomix [~gsomix@188.168.4.249] has joined #shogun | 05:11 | |
-!- muddo [~muddo@gateway/tor-sasl/muddo] has quit [Ping timeout: 276 seconds] | 05:26 | |
-!- vikram360 [~vikram360@117.192.171.206] has joined #shogun | 05:39 | |
-!- harshit_ [~harshit@182.68.171.83] has joined #shogun | 05:59 | |
-!- harshit_ [~harshit@182.68.171.83] has quit [Ping timeout: 260 seconds] | 06:05 | |
-!- harshit_ [~harshit@182.68.171.83] has joined #shogun | 06:43 | |
-!- gsomix [~gsomix@188.168.4.249] has quit [Ping timeout: 260 seconds] | 06:51 | |
-!- harshit_ [~harshit@182.68.171.83] has quit [Read error: Connection reset by peer] | 07:33 | |
-!- harshit_ [~harshit@182.68.32.27] has joined #shogun | 07:36 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 07:36 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Client Quit] | 07:39 | |
-!- harshit_ [~harshit@182.68.32.27] has quit [Ping timeout: 264 seconds] | 07:41 | |
-!- harshit_ [~harshit@182.68.32.27] has joined #shogun | 07:41 | |
-!- hpsoar [7c108901@gateway/web/freenode/ip.124.16.137.1] has joined #shogun | 07:43 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 07:46 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 07:53 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 07:58 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 08:10 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Client Quit] | 08:11 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 08:13 | |
-!- puneetgoyal [~puneet@115.241.145.47] has joined #shogun | 08:16 | |
-!- novice [b4953264@gateway/web/freenode/ip.180.149.50.100] has joined #shogun | 08:19 | |
-!- novice [b4953264@gateway/web/freenode/ip.180.149.50.100] has quit [Client Quit] | 08:21 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Quit: wiking] | 08:31 | |
-!- puneetgoyal [~puneet@115.241.145.47] has quit [Ping timeout: 245 seconds] | 08:41 | |
hpsoar | hi, my github id is hpsoar | 08:48 |
---|---|---|
hpsoar | today, I compiled and installed shogun on my notebook with linux os | 08:49 |
hpsoar | just now, i set up my github tools, and successfully sent a pull request test | 08:49 |
hpsoar | I had encountered some problems when install shogun, and run the examples | 08:50 |
-!- harshit_ [~harshit@182.68.32.27] has quit [Ping timeout: 250 seconds] | 08:51 | |
-!- puneetgoyal [~puneet@115.240.54.109] has joined #shogun | 08:52 | |
hpsoar | 1. firstly, I install through apt-get, but the examples doesn't work with errors like "cannot import name AveragedPerceptron" | 08:52 |
n4nd0 | hpsoar: hey! so the pacakge is not really up to date | 08:54 |
n4nd0 | hpsoar: you should try to compile and install from source, the code in github | 08:55 |
hpsoar | yes, I reinstalled through the source code | 08:55 |
hpsoar | the examples can run, but some ends up unconverged | 08:56 |
n4nd0 | hpsoar: I am not sure if I follow | 08:56 |
n4nd0 | hpsoar: so it seems that you are trying out python examples, did you install with python support? | 08:57 |
hpsoar | yes | 08:57 |
n4nd0 | ./configure --interfaces=python_modular ?? | 08:57 |
hpsoar | yes | 08:57 |
n4nd0 | ok | 08:57 |
n4nd0 | and you are able or not to run examples? | 08:57 |
hpsoar | yes | 08:57 |
n4nd0 | then everything is good :) | 08:58 |
hpsoar | ok, I want participate in the gsoc project, I think I need to spend some time getting familiar with the project. thanks for your help~ | 09:02 |
-!- harshit_ [~harshit@182.68.120.246] has joined #shogun | 09:03 | |
-!- muddo [~muddo@gateway/tor-sasl/muddo] has joined #shogun | 09:04 | |
n4nd0 | no problem, just ask around here if you have any problem so we can try to help you ;) | 09:05 |
-!- puneetgoyal [~puneet@115.240.54.109] has quit [Ping timeout: 245 seconds] | 09:25 | |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has joined #shogun | 09:26 | |
blackburn | hpsoar: it seems we've got some misunderstanding at pull request point | 09:27 |
blackburn | have you subscribed to the mailing list? | 09:27 |
hpsoar | yes | 09:27 |
blackburn | hpsoar: I clarified it a little, we intent that students could help with some improvements | 09:28 |
blackburn | but not some test thing | 09:28 |
hpsoar | doesn't the project require a pull request test to make sure that the version control can work correctly | 09:28 |
hpsoar | I know that, I have already viewed the ideas proposed | 09:29 |
blackburn | no, github works well anyway | 09:30 |
blackburn | we suggest students to make real pull requests | 09:30 |
blackburn | that's what I mean | 09:30 |
blackburn | hpsoar: according to things you said before (I checked logs) - contact mentor(s) of idea(s) you are interested with | 09:34 |
hpsoar | oh, sorry | 09:35 |
blackburn | no problem | 09:35 |
hpsoar | I haved already pull the source to my machine, ok, thank you | 09:35 |
-!- hpsoar [7c108901@gateway/web/freenode/ip.124.16.137.1] has left #shogun [] | 09:41 | |
sonne|work | blackburn: thx for the clarification email | 09:44 |
blackburn | sonne|work: lol | 09:44 |
sonne|work | btw - I think I know what is wrong with tests | 09:44 |
blackburn | what? | 09:44 |
sonne|work | serialization seems to not work | 09:44 |
blackburn | oh awesome | 09:44 |
sonne|work | no idea why though | 09:44 |
sonne|work | pickle.dumps()/loads() work... | 09:44 |
blackburn | sonne|work: I am tending to abandon release right now | 09:45 |
sonne|work | so all tests fail that serialize shogun objects | 09:45 |
sonne|work | ? | 09:45 |
blackburn | well we may not release this time I mean | 09:45 |
sonne|work | if I figure out what is wrong - I think we are in good shape for a release | 09:45 |
blackburn | ok, I don't mind release but I mean I hardly can find much time to fix some serialization errors | 09:46 |
n4nd0 | if I can help with sth guys, let me know please | 09:47 |
sonne|work | blackburn: it is clear that I have to figure out what is wrong... - well I can try | 09:48 |
-!- flxb [~cronor@fb.ml.tu-berlin.de] has joined #shogun | 09:48 | |
blackburn | sonne|work: not so clear if you have no time for that :) | 09:49 |
flxb | Hello! | 09:49 |
flxb | Is it possible to serialize a GMNPSVM classifier? And if so, how do you do it in python_modular? | 09:49 |
sonne|work | n4nd0: it is not so easy - I think I have to do it this time | 09:50 |
sonne|work | pickle.dump(classifier) | 09:50 |
blackburn | yes it is possible | 09:50 |
n4nd0 | sonne|work: ok | 09:50 |
blackburn | n4nd0: what about multiclassmachine? | 09:55 |
blackburn | I was lazy to implement OvO training | 09:55 |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has left #shogun [] | 09:57 | |
n4nd0 | blackburn: do you mean to prepare an interface similar to LinearMachine or the like for multiclass? | 09:57 |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has joined #shogun | 09:57 | |
blackburn | wrong click lol | 09:58 |
n4nd0 | blackburn: I don't know if you get it before | 09:58 |
n4nd0 | blackburn: do you mean to prepare an interface similar to LinearMachine or the like for multiclass? | 09:58 |
blackburn | n4nd0: check shogun/machine/MulticlassMachine.cpp | 09:58 |
blackburn | n4nd0: there is a code training machines with OvR scheme | 09:58 |
blackburn | so approach is similar | 09:59 |
blackburn | but you need to subset features instance: i.e. select only feature vectors that belongs to 2 classes you dichotomize right now | 09:59 |
blackburn | LibSVMMultiClass has some that kind of code | 10:00 |
blackburn | and MultiClassSVM class has apply for that scheme | 10:00 |
n4nd0 | blackburn: 0vR is the same as one vs. all? | 10:01 |
blackburn | yeah | 10:01 |
blackburn | I used to say OvR cause 'all' sounds like you try to construct hyperplane between set X and set X+Y - this makes no sense :) | 10:02 |
n4nd0 | haha ok, I see your porint One vs. Rest | 10:02 |
n4nd0 | point* | 10:02 |
n4nd0 | so this 0v0 scheme is implemented in LibSVM but not in shogun? | 10:04 |
blackburn | I want to port all multiclass things into multiclass folder | 10:05 |
blackburn | GMNPSVM in fact | 10:06 |
blackburn | and LibSVMMulticlass as well | 10:06 |
blackburn | let me check how it trains | 10:06 |
blackburn | sonne|work: about labels I postpone it to some day after release | 10:07 |
blackburn | binarylabels, multiclasslabels, probabilisticlabels, etc | 10:07 |
blackburn | n4nd0: ok multiclass libsvm trains 1-vs-1 | 10:10 |
n4nd0 | n4nd0: all right, and that's already in shogun | 10:11 |
blackburn | yes, but it should change its base | 10:11 |
n4nd0 | blackburn: how should it be? | 10:12 |
blackburn | I introduced MulticlassMachine | 10:12 |
blackburn | probably MulticlassSVM should be derived from KernelMulticlassMachine | 10:12 |
blackburn | and all these things should -> multiclass instead of shogun/classifier/svm | 10:13 |
n4nd0 | blackburn: got it | 10:14 |
blackburn | n4nd0: but this time you probably better implement that generic thing | 10:14 |
blackburn | in multiclassmachine | 10:14 |
blackburn | do you know how 1-vs-1 works? | 10:14 |
n4nd0 | blackburn: what is that generic thing? | 10:15 |
n4nd0 | blackburn: no, I have just heard of 1vsR | 10:15 |
blackburn | both LinearMulticlassMachine and KernelMulticlassMachine can use any classifier | 10:15 |
blackburn | linear or kernel based accordingly | 10:15 |
blackburn | i.e. you may | 10:16 |
blackburn | LinearMulticlassMachine(ONE_VS_REST_STRATEGY, QDA(), train_features, train_labels) | 10:16 |
blackburn | ah yes QDA is not linera | 10:16 |
blackburn | then LDA() | 10:16 |
n4nd0 | :) | 10:16 |
blackburn | so when you train it | 10:17 |
blackburn | it constructs 1-vs-rest linear classifiers using provided lda | 10:17 |
blackburn | and then classifies using these constructed hyperplanes :) | 10:17 |
n4nd0 | I think I get the idea | 10:18 |
n4nd0 | I guess this 1vs1 is another strategy of doing multiclass instead os 1vsR | 10:18 |
blackburn | yes | 10:19 |
blackburn | take a look no train_one_vs_rest | 10:19 |
blackburn | imagine you've got labels [0,0,0,1,1,1,2,2,2] | 10:19 |
blackburn | it constructs labels | 10:19 |
blackburn | [-1,-1,-1,+1,+1,+1,+1,+1,+1] | 10:19 |
blackburn | [+1,+1,+1,-1,-1,-1,+1,+1,+1] | 10:19 |
blackburn | [+1,+1,+1,+1,+1,+1,-1,-1,-1] | 10:20 |
n4nd0 | I see | 10:20 |
blackburn | and trains provided machine using ^ | 10:20 |
blackburn | so 3 planes separating each class from rest | 10:20 |
blackburn | train_one_vs_one should be a little different | 10:20 |
blackburn | imagine again you've got labels [0,0,0,1,1,1,2,2,2] | 10:21 |
blackburn | [-1,-1,-1,+1,+1,+1,0,0,0] | 10:21 |
blackburn | [-1,-1,-1,0,0,0,+1,+1,+1] | 10:21 |
blackburn | [0,0,0,-1,-1,-1,+1,+1,+1] | 10:22 |
blackburn | we should exclude vectors with 0 label | 10:22 |
blackburn | they are not considered by machine at all | 10:22 |
blackburn | got it? | 10:22 |
n4nd0 | yes | 10:22 |
blackburn | this can be done with subsetting | 10:23 |
n4nd0 | in a general case 1vs1 generates more separating hyperplanes right? | 10:23 |
blackburn | if there are K classes | 10:24 |
blackburn | it constructs K*(K-1)/2 | 10:24 |
n4nd0 | while the other always does K | 10:24 |
blackburn | OvR constructs K | 10:25 |
blackburn | yes | 10:25 |
n4nd0 | good | 10:25 |
n4nd0 | so the idea is first to move all the stuff to multiclass | 10:25 |
n4nd0 | and later implement this one vs. one | 10:25 |
blackburn | no | 10:26 |
blackburn | just implement one vs one | 10:26 |
blackburn | in case Soeren would manage to repair serialization | 10:27 |
blackburn | we release, so no need to do such intrusive changes | 10:27 |
n4nd0 | ok | 10:27 |
blackburn | n4nd0: btw there should be for i=1:K for j=i+1:K | 10:28 |
blackburn | little hint :) | 10:29 |
blackburn | i.e. no need to construct classifiers between both (0,1) and (1,0) | 10:29 |
blackburn | they are equal up to sign | 10:29 |
n4nd0 | :) | 10:29 |
n4nd0 | good you said, I might have repeated stuff | 10:30 |
blackburn | for i = 1 : K | 10:30 |
blackburn | for j = i + 1 : K | 10:30 |
n4nd0 | so simplying the idea is to add two methods in MulticlassMachine | 10:31 |
blackburn | mark labels == i as -1 | 10:31 |
blackburn | mark labels == j as +1 | 10:31 |
blackburn | subset feature vectors that belong to i and j classes (exclude others) | 10:31 |
blackburn | train classifier, store it | 10:31 |
blackburn | that's THE MARVELLOUS algorithm ;) | 10:31 |
n4nd0 | :) | 10:32 |
blackburn | n4nd0: yes, train_one_vs_one | 10:32 |
blackburn | apply_one_vs_one | 10:32 |
n4nd0 | ok | 10:32 |
blackburn | apply_one_vs_one is implemented in MultiClassSVM | 10:32 |
blackburn | just copy it | 10:32 |
-!- harshit_ [~harshit@182.68.120.246] has quit [Ping timeout: 246 seconds] | 10:32 | |
n4nd0 | blackburn: ok, it doesn't look like much work then (one never knows though) | 10:33 |
blackburn | n4nd0: yes it is pretty simpel | 10:33 |
blackburn | :) | 10:33 |
blackburn | I did not implement it cause I am having some research on task with 43 classes | 10:33 |
n4nd0 | blackburn: I will try to do it and test it, comparing it to another example using 1vsR and hope to get same results | 10:33 |
blackburn | so damn hard to train 903 classifiers | 10:33 |
blackburn | applying with libsvmmulticlass takes ONE HOUR | 10:34 |
blackburn | thanks to wiking we've got a way to make it linear | 10:34 |
blackburn | now it takes 100s | 10:34 |
blackburn | :D | 10:34 |
blackburn | to train* | 10:34 |
blackburn | and 4s to apply | 10:34 |
n4nd0 | cool | 10:35 |
blackburn | n4nd0: ok I probably should quit irc cause it is pretty distracting and I can't start working :) | 10:44 |
blackburn | to get fired during this crazy pre-gsoc is not my intention :D | 10:44 |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 10:45 | |
blackburn | ah nevermind I'll just stop staring | 10:46 |
blackburn | :D | 10:46 |
blackburn | highlight me if needed | 10:47 |
n4nd0 | ok, will do that | 10:50 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 10:50 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 10:50 | |
-!- in3xes_ [~in3xes@180.149.49.227] has joined #shogun | 11:29 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 252 seconds] | 11:33 | |
-!- GenX [~Sambhav@14.139.82.6] has quit [Read error: Connection reset by peer] | 11:33 | |
-!- GenX [~Sambhav@14.139.82.6] has joined #shogun | 11:33 | |
-!- harshit_ [~harshit@182.68.120.246] has joined #shogun | 11:34 | |
-!- GenX__ [~Sambhav@14.139.82.6] has joined #shogun | 11:35 | |
-!- GenX [~Sambhav@14.139.82.6] has quit [Ping timeout: 244 seconds] | 11:38 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 244 seconds] | 11:39 | |
-!- gsomix [~gsomix@188.168.4.113] has joined #shogun | 11:44 | |
-!- gsomix [~gsomix@188.168.4.113] has quit [Ping timeout: 246 seconds] | 11:53 | |
-!- in3xes_ [~in3xes@180.149.49.227] has quit [Ping timeout: 246 seconds] | 11:58 | |
-!- gsomix [~gsomix@188.168.14.102] has joined #shogun | 12:02 | |
harshit_ | is there any function in Shogun or in any c++ library to check whether matrix is sparse or not | 12:54 |
harshit_ | ? | 12:54 |
-!- hpsoar [7c108901@gateway/web/freenode/ip.124.16.137.1] has joined #shogun | 12:55 | |
-!- harshit_ [~harshit@182.68.120.246] has quit [Ping timeout: 264 seconds] | 13:11 | |
-!- hpsoar [7c108901@gateway/web/freenode/ip.124.16.137.1] has quit [Quit: Page closed] | 13:17 | |
-!- novice [b4953263@gateway/web/freenode/ip.180.149.50.99] has joined #shogun | 13:24 | |
-!- novice [b4953263@gateway/web/freenode/ip.180.149.50.99] has quit [Client Quit] | 13:26 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 13:31 | |
-!- wangxiang [~chatzilla@125.33.178.131] has joined #shogun | 13:35 | |
-!- wangxiang [~chatzilla@125.33.178.131] has quit [Quit: ChatZilla 0.9.88.1 [Firefox 9.0.1/20111220165912]] | 13:43 | |
-!- l0nr4n [~l0nr4n@unaffiliated/l0nr4n] has joined #shogun | 13:52 | |
-!- n4nd0 [~fjig@grey-02.csc.kth.se] has joined #shogun | 14:01 | |
-!- in3xes [~in3xes@59.163.196.121] has joined #shogun | 14:15 | |
-!- hpsoar [7c108901@gateway/web/freenode/ip.124.16.137.1] has joined #shogun | 14:19 | |
hpsoar | why the 1.2.0 version can not configure with cmdline_modular? | 14:20 |
blackburn | cmdline_modular? there is no such interface | 14:21 |
wiking | yo | 14:22 |
blackburn | hey | 14:22 |
wiking | wazza man | 14:26 |
wiking | finally i'm work mode again | 14:26 |
wiking | instead of writing paper mode | 14:26 |
n4nd0 | wiking: what was your paper about? | 14:27 |
wiking | n4nd0: image similarity measures | 14:27 |
wiking | basically extending my prev algorithms on computer vision for getting better accuracy | 14:27 |
wiking | nothing fancy :| | 14:27 |
wiking | it's one of those need-to-publish papers in my opinion | 14:28 |
n4nd0 | wiking: aham hope you get a good publication anyway :) | 14:28 |
wiking | heheh let's hope so | 14:28 |
-!- vikram360 [~vikram360@117.192.171.206] has quit [Ping timeout: 260 seconds] | 14:34 | |
blackburn | wiking: n4nd0 guys last year we had some troubles with git, I just noticed you have no troubles at all :) | 14:43 |
n4nd0 | :) | 14:45 |
n4nd0 | let's hope it continues like this, without problems | 14:45 |
wiking | blackburn: ? | 14:49 |
blackburn | wiking: well we had some issues with branching etc, we all were new to git :D | 14:49 |
n4nd0 | svn before? | 14:50 |
wiking | hahahah ah ok | 14:51 |
wiking | :) | 14:51 |
blackburn | actually there are some pitfalls | 14:52 |
blackburn | e.g. you may see Merge branch ... " commits | 14:52 |
blackburn | sometimes | 14:52 |
blackburn | they can be avoided with proper rebasing | 14:52 |
-!- flxb [~cronor@fb.ml.tu-berlin.de] has quit [Ping timeout: 245 seconds] | 14:52 | |
-!- vikram360 [~vikram360@117.216.144.158] has joined #shogun | 15:02 | |
-!- in3xes [~in3xes@59.163.196.121] has quit [Quit: Leaving] | 15:03 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 15:39 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Remote host closed the connection] | 15:43 | |
-!- hpsoar [7c108901@gateway/web/freenode/ip.124.16.137.1] has quit [Ping timeout: 245 seconds] | 15:47 | |
wiking | heheh yeah rebasing is gooood | 15:51 |
wiking | blackburn: branching and merging is like amaaaazing compared to svn ...! | 16:04 |
blackburn | you tend to make words long today ;) | 16:05 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Remote host closed the connection] | 16:05 | |
-!- wiking [~wiking@huwico/staff/wiking] has joined #shogun | 16:05 | |
-!- wiking_ [~wiking@huwico/staff/wiking] has joined #shogun | 16:07 | |
blackburn | wiking: one more wiking and you all will break the ice and sink | 16:08 |
-!- wiking__ [~wiking@huwico/staff/wiking] has joined #shogun | 16:08 | |
-!- wiking_ [~wiking@huwico/staff/wiking] has quit [Read error: Connection reset by peer] | 16:08 | |
blackburn | oh no | 16:08 |
blackburn | one sinked | 16:08 |
-!- wiking__ [~wiking@huwico/staff/wiking] has quit [Remote host closed the connection] | 16:08 | |
blackburn | another one bites the dust | 16:09 |
-!- wiking_ [~wiking@huwico/staff/wiking] has joined #shogun | 16:09 | |
blackburn | and another one comes | 16:09 |
wiking_ | hahahah | 16:09 |
wiking_ | oh sorry to be proper.... aaaaaahhahahaa | 16:09 |
blackburn | ;) | 16:09 |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 265 seconds] | 16:11 | |
-!- wiking_ is now known as wiking | 16:11 | |
-!- blackburn [5bdfb203@gateway/web/freenode/ip.91.223.178.3] has quit [Quit: Page closed] | 16:26 | |
-!- n4nd0 [~fjig@grey-02.csc.kth.se] has quit [Quit: leaving] | 16:26 | |
-!- in3xes [~in3xes@59.163.196.5] has joined #shogun | 16:48 | |
-!- vikram360 [~vikram360@117.216.144.158] has quit [Ping timeout: 245 seconds] | 16:49 | |
-!- in3xes_ [~in3xes@210.212.58.168] has joined #shogun | 16:51 | |
-!- in3xes [~in3xes@59.163.196.5] has quit [Ping timeout: 260 seconds] | 16:55 | |
-!- gsomix [~gsomix@188.168.14.102] has quit [Read error: Operation timed out] | 16:56 | |
-!- in3xes_ is now known as in3xes | 16:57 | |
-!- vikram360 [~vikram360@117.216.144.158] has joined #shogun | 17:06 | |
-!- vikram360 [~vikram360@117.216.144.158] has quit [Ping timeout: 252 seconds] | 17:27 | |
-!- sleepingdragon [ca4eafc7@gateway/web/freenode/ip.202.78.175.199] has joined #shogun | 17:40 | |
-!- sleepingdragon [ca4eafc7@gateway/web/freenode/ip.202.78.175.199] has quit [Client Quit] | 17:41 | |
-!- vikram360 [~vikram360@117.192.161.75] has joined #shogun | 17:59 | |
-!- wiking [~wiking@huwico/staff/wiking] has quit [Ping timeout: 248 seconds] | 18:00 | |
-!- vikram360 [~vikram360@117.192.161.75] has quit [Ping timeout: 244 seconds] | 18:05 | |
-!- vikram360 [~vikram360@117.192.187.71] has joined #shogun | 18:06 | |
-!- gsomix [~gsomix@83.234.54.97] has joined #shogun | 18:38 | |
-!- harshit_ [~harshit@182.68.120.246] has joined #shogun | 18:53 | |
-!- vikram360 [~vikram360@117.192.187.71] has quit [Ping timeout: 276 seconds] | 19:14 | |
-!- harshit_ [~harshit@182.68.120.246] has quit [Ping timeout: 260 seconds] | 19:18 | |
-!- vikram360 [~vikram360@117.192.187.71] has joined #shogun | 19:29 | |
-!- in3xes_ [~in3xes@59.163.196.5] has joined #shogun | 19:32 | |
-!- harshit_ [~harshit@182.68.120.246] has joined #shogun | 19:37 | |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 20:00 | |
-!- vikram360 [~vikram360@117.192.187.71] has quit [Ping timeout: 260 seconds] | 20:06 | |
CIA-64 | shogun: Soeren Sonnenburg master * r5efddaa / src/shogun/evaluation/CrossValidation.cpp : fix spelling mistake -> warning - http://git.io/xTbBTg | 20:10 |
CIA-64 | shogun: Soeren Sonnenburg master * r7bb576d / (2 files in 2 dirs): extract mahalanobis description and don't print anything in example - http://git.io/sC6bKg | 20:10 |
CIA-64 | shogun: Soeren Sonnenburg master * r7329de6 / testsuite/python_modular/tester.py : compare pickled version of SGObject instances not just ptrs - http://git.io/OS3iVg | 20:10 |
-!- muddo [~muddo@gateway/tor-sasl/muddo] has quit [Ping timeout: 276 seconds] | 20:25 | |
-!- puneetgoyal [~puneet@115.240.121.85] has joined #shogun | 20:28 | |
-!- muddo [~muddo@gateway/tor-sasl/muddo] has joined #shogun | 20:34 | |
-!- vikram360 [~vikram360@117.192.187.71] has joined #shogun | 20:45 | |
CIA-64 | shogun: Soeren Sonnenburg master * rd4bb942 / applications/vw/Makefile : fix vw makefile - http://git.io/Bw5r8A | 21:02 |
CIA-64 | shogun: Soeren Sonnenburg master * reca0d2d / (4 files in 4 dirs): | 21:02 |
CIA-64 | shogun: fix a couple of issues like in applications | 21:02 |
CIA-64 | shogun: most notably the SVM object is gone and replaced by KernelMachine - http://git.io/XTq-mQ | 21:02 |
-!- gsomix [~gsomix@83.234.54.97] has quit [Quit: ????? ? ?? ??? (xchat 2.4.5 ??? ??????)] | 21:16 | |
-!- blackburn [~qdrgsm@188.168.4.179] has joined #shogun | 21:16 | |
-!- l0nr4n [~l0nr4n@unaffiliated/l0nr4n] has quit [Quit: This computer (or maybe me) has gone to sleep] | 21:21 | |
-!- novice [b4953263@gateway/web/freenode/ip.180.149.50.99] has joined #shogun | 21:24 | |
-!- novice [b4953263@gateway/web/freenode/ip.180.149.50.99] has quit [Client Quit] | 21:24 | |
-!- PhilTillet [9d9f0a0e@gateway/web/freenode/ip.157.159.10.14] has joined #shogun | 21:25 | |
PhilTillet | Hello | 21:26 |
-!- harshit_ [~harshit@182.68.120.246] has quit [Quit: Leaving] | 21:31 | |
@sonney2k | PhilTillet, hi | 21:32 |
blackburn | hey guys | 21:32 |
blackburn | sonney2k: did you manage to fix tester? | 21:32 |
-!- PhilTillet [9d9f0a0e@gateway/web/freenode/ip.157.159.10.14] has quit [Ping timeout: 245 seconds] | 21:32 | |
blackburn | bam | 21:32 |
blackburn | another one bites the dust | 21:32 |
-!- flxb [~cronor@g230122103.adsl.alicedsl.de] has joined #shogun | 21:34 | |
blackburn | sonney2k: I mean I have seen you fixed something | 21:34 |
CIA-64 | shogun: Soeren Sonnenburg master * rf081ae0 / testsuite/python_modular/tester.py : only import modshogun - http://git.io/dkMc5Q | 21:34 |
CIA-64 | shogun: Soeren Sonnenburg master * raba8078 / (4 files): comment print functions in examples - http://git.io/mnNbyw | 21:34 |
blackburn | and again | 21:35 |
@sonney2k | blackburn, do you know whether heiko will be around? | 21:35 |
blackburn | sonney2k: he said this week is pretty hard | 21:35 |
-!- in3xes [~in3xes@210.212.58.168] has quit [Ping timeout: 240 seconds] | 21:35 | |
blackburn | so probably next week only | 21:35 |
@sonney2k | I think there is some breakage in this data lock stuff / at least kernels etc are no longer set in gmnpsvm after de-serialization | 21:36 |
blackburn | sonney2k: row subset thing? | 21:36 |
@sonney2k | no | 21:36 |
-!- in3xes_ [~in3xes@59.163.196.5] has quit [Ping timeout: 272 seconds] | 21:36 | |
@sonney2k | maybe related - I don't know for sure | 21:36 |
blackburn | sonney2k: actually I asked him to fix this parameters thing | 21:36 |
@sonney2k | basically all applications are broken now | 21:37 |
@sonney2k | but tester.py -t 1e-10 runs many examples fine now | 21:37 |
blackburn | sonney2k: I probably know the error - we need to add new field to parameter map | 21:37 |
@sonney2k | so some improvements but not down to 0 errors | 21:38 |
blackburn | I see | 21:38 |
-!- flxb [~cronor@g230122103.adsl.alicedsl.de] has left #shogun [] | 21:38 | |
blackburn | sonney2k: I'll try to fix it but tomorrow or later | 21:38 |
blackburn | have to go back to studies again damn | 21:38 |
blackburn | sonney2k: time to study signal processing! see you ;) | 21:40 |
n4nd0 | blackburn: signal processing, cool stuff :P | 21:40 |
blackburn | n4nd0: I am not a big fan but you are probably right ;) | 21:41 |
blackburn | n4nd0: so see you too :D | 21:41 |
blackburn | bye | 21:41 |
n4nd0 | :) bye! | 21:41 |
n4nd0 | sonney2k: nico answered today, you were right :) | 21:42 |
-!- blackburn [~qdrgsm@188.168.4.179] has quit [Read error: Operation timed out] | 21:43 | |
-!- in3xes [~in3xes@180.149.49.227] has joined #shogun | 21:46 | |
-!- gsomix [~gsomix@83.234.54.97] has joined #shogun | 21:50 | |
-!- vikram360 [~vikram360@117.192.187.71] has quit [Ping timeout: 265 seconds] | 21:56 | |
-!- aang [0e8be4d2@gateway/web/freenode/ip.14.139.228.210] has joined #shogun | 22:33 | |
-!- puneetgoyal [~puneet@115.240.121.85] has quit [Ping timeout: 240 seconds] | 22:44 | |
-!- muddo [~muddo@gateway/tor-sasl/muddo] has quit [Ping timeout: 276 seconds] | 22:46 | |
-!- muddo [~muddo@gateway/tor-sasl/muddo] has joined #shogun | 22:57 | |
-!- PhilTillet [9d9f0a0e@gateway/web/freenode/ip.157.159.10.14] has joined #shogun | 23:08 | |
PhilTillet | Nobody's insomniac today :D | 23:13 |
-!- GenX__ is now known as GenX | 23:14 | |
n4nd0 | PhilTillet: there's always someone :P | 23:17 |
PhilTillet | Haha :D | 23:18 |
-!- gsomix [~gsomix@83.234.54.97] has quit [Ping timeout: 246 seconds] | 23:25 | |
n4nd0 | but time to go now | 23:31 |
n4nd0 | good night guys | 23:31 |
-!- n4nd0 [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 23:32 | |
-!- in3xes [~in3xes@180.149.49.227] has quit [Ping timeout: 260 seconds] | 23:39 | |
-!- l0nr4n [~l0nr4n@g231201109.adsl.alicedsl.de] has joined #shogun | 23:53 | |
-!- l0nr4n [~l0nr4n@g231201109.adsl.alicedsl.de] has quit [Changing host] | 23:53 | |
-!- l0nr4n [~l0nr4n@unaffiliated/l0nr4n] has joined #shogun | 23:53 | |
--- Log closed Tue Mar 20 00:00:19 2012 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!