| --- Log opened Sat Aug 03 00:00:59 2013 | ||
| -!- pickle27 [~Kevin@d67-193-243-174.home3.cgocable.net] has quit [Quit: Leaving] | 00:05 | |
| -!- travis-ci [~travis-ci@ec2-54-226-85-18.compute-1.amazonaws.com] has joined #shogun | 00:25 | |
| travis-ci | [travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/9792052 | 00:25 |
|---|---|---|
| -!- travis-ci [~travis-ci@ec2-54-226-85-18.compute-1.amazonaws.com] has left #shogun [] | 00:25 | |
| hushell | iglesiasg: hi | 00:42 |
| @iglesiasg | hey hushell! | 00:42 |
| @iglesiasg | how are you doing_ | 00:42 |
| @iglesiasg | ? | 00:42 |
| hushell | iglesiasg: I am implementing code after all meetings with my supervisors :) | 00:42 |
| hushell | iglesiasg: A strange problem | 00:43 |
| @iglesiasg | hushell, lot of work in related to the phd these days? | 00:43 |
| @iglesiasg | what is it? | 00:43 |
| hushell | iglesiasg: undefined reference to `shogun::SGVector<shogun::CFactor*>::~SGVector()' | 00:43 |
| hushell | I was trying to use SGVector<CSGObject*> also failed, only int float supported | 00:44 |
| @iglesiasg | hushell, yes | 00:44 |
| hushell | iglesiasg: yeah, quit busy these days, too many meetings, but I'll be free next week :) | 00:44 |
| @iglesiasg | hushell, look at SGVector.cpp | 00:44 |
| @iglesiasg | hushell, the last lines. Those are the types that SGVector is usable for | 00:45 |
| hushell | but I checked base/Parameters.h there is SGVector<CSGObject*> | 00:45 |
| @iglesiasg | aham, then you already know more than I do about that hehe | 00:45 |
| hushell | iglesiasg: okay, I see I was strange why template functions can be declared in different files | 00:46 |
| hushell | iglesiasg: but SGVector<CSGObject*> cannot be used | 00:47 |
| @iglesiasg | hushell, yeah. It is a trick to separate both files | 00:47 |
| @iglesiasg | hushell, well if it is used in Parameter.h, it should be possible! | 00:47 |
| hushell | iglesiasg: okay which means I should resolve DynamicObjectArray or CList somthing like those? | 00:48 |
| hushell | iglesiasg: resort to I mean | 00:48 |
| @iglesiasg | hushell, I think I have never used SGVector for things other than primitive types | 00:49 |
| @iglesiasg | hushell, but if it appears in Parameter.h, so maybe it is possible | 00:49 |
| hushell | iglesiasg: what will you do if you need to store a bunch of class instances | 00:49 |
| @iglesiasg | hushell, it depends what do I need to do with them | 00:50 |
| hushell | std::vector is the best way but I think it is not allowed | 00:50 |
| @iglesiasg | hushell, std::vector is fine as long as you can avoid including it in .h files that are visible to SWIG | 00:50 |
| hushell | iglesiasg: in my case, I have a sequence of messages for each edge in the graph | 00:50 |
| hushell | I need to store them | 00:51 |
| shogun-buildbot | build #1146 of cyg1 - libshogun is complete: Failure [failed configure] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1146 blamelist: Soeren Sonnenburg <sonne@debian.org> | 00:51 |
| @iglesiasg | this is, if it is something internal of the implementation, and not something that is going to be exposed in the interface. then it is completely fine to use STL | 00:51 |
| hushell | iglesiasg: but I may need keep it as members | 00:51 |
| @iglesiasg | hushell, then I guess DynamicObjectArray is the way to go | 00:53 |
| @iglesiasg | but I am curious about this use of SGVector in Parameter.h | 00:53 |
| hushell | iglesiasg: I tried to include SGObject.h and use SGVector<CSGObject*>, not working | 00:53 |
| @iglesiasg | hushell, yeah, it doesn't work for me either | 00:54 |
| @iglesiasg | simple example declaring | 00:54 |
| @iglesiasg | SGVector<CSGObject> vector(5); | 00:54 |
| @iglesiasg | compile error due to undefined reference | 00:54 |
| hushell | iglesiasg: DynamicObjectArray is not very convenient honestly | 00:55 |
| @iglesiasg | hushell, what is wrong with it? | 00:55 |
| @iglesiasg | IIRC is sort of std::vector | 00:55 |
| @iglesiasg | you have got a similar API with push_back and so | 00:55 |
| hushell | iglesiasg: yeah, it's efficient but I don't like the type conversion code | 00:56 |
| @iglesiasg | hushell, aham, that some casts have to be made explicitly? | 00:57 |
| hushell | iglesiasg: I found some code has std::vector as members :D | 00:57 |
| @iglesiasg | hushell, too bad. That is not usable from SWIG I am afraid. Where? | 00:57 |
| @iglesiasg | maybe it is a class that is not visible to SWIG | 00:58 |
| @iglesiasg | note that all of them are not, one has to make that explicit | 00:58 |
| hushell | iglesiasg: maybe it's not big case, e.g. shogun/multiclass/LaRank.h:295 | 01:01 |
| hushell | iglesiasg: you can use this command to search it: find . -type f -exec grep -l 'std::' {} \; | 01:01 |
| @iglesiasg | hushell, cool thanks for little script! practical indeed | 01:02 |
| @iglesiasg | hushell, I bet that is sort of an internal class | 01:03 |
| @iglesiasg | hushell, grep -R LaRankPatterns * in src/interfaces | 01:03 |
| hushell | I think so | 01:04 |
| @iglesiasg | hushell, that class is ignored by SWIG, so it is fine to use STL | 01:04 |
| hushell | iglesiasg: if a class is going to be used in other languages, no STL in function arguments and class members, right? | 01:05 |
| @iglesiasg | hushell, exactly | 01:05 |
| hushell | iglesiasg: I can put these message stuff inside a function, but maybe not quite efficient, they have to be recomputed :( | 01:07 |
| @iglesiasg | hushell, mmm | 01:08 |
| hushell | iglesiasg: okay, I will use DynamicObjectArray ... | 01:08 |
| hushell | and make message class be a derived class from SGObject | 01:09 |
| @iglesiasg | hushell, why don't you like the casts? performance, syntax, another thing? | 01:09 |
| hushell | iglesiasg: :D too much typing | 01:10 |
| @iglesiasg | hushell, haha ok | 01:10 |
| hushell | iglesiasg: Thanks, I guess it's very late in your time | 01:14 |
| @iglesiasg | hushell, 1:15am | 01:15 |
| @iglesiasg | hushell, but I am sort of a night guy hehe | 01:15 |
| @iglesiasg | hushell, so how is the inference going? | 01:21 |
| hushell | iglesiasg: I am also late person :) you always online when I want to find someone to talk | 01:22 |
| hushell | I'll be finish it tonight, so you will see when you get up | 01:22 |
| hushell | I have commited half to my github | 01:22 |
| hushell | maybe I can send a PR now, 690 lines already | 01:23 |
| @iglesiasg | hushell, it sounds good! | 01:25 |
| @iglesiasg | but no rush, just issue the PR when you think it is ready | 01:25 |
| hushell | okay, because no documentation there I think do it later better | 01:26 |
| @iglesiasg | hushell, I am working on LMNN | 01:31 |
| @iglesiasg | hushell, it is not scalable *at all* at this moment :( | 01:31 |
| hushell | iglesiasg: good, all right, I go back to work | 01:35 |
| hushell | iglesiasg: hmm, I guess maybe the problem is the algorithm itself not your code | 01:36 |
| @iglesiasg | hushell, unfortunately not, I am comparing with the author's implementation. It can handle bigger datasets than mine | 01:37 |
| @iglesiasg | hushell, but it is fine. I just made a couple of big simplifications at the beginning | 01:37 |
| @iglesiasg | hushell, I am refining at the moment. The one I am implementing now worked pretty well in my Matlab prototype so I have big hopes for it | 01:37 |
| hushell | iglesiasg: interesting, no worries, we got time. what will be your demo? I'll let my college to use it after gsoc | 01:40 |
| @iglesiasg | hushell, classification in metagenomics data | 01:40 |
| hushell | I am trying to have some image demos, but Patrick likes OCR better, something like POS tagging | 01:45 |
| @iglesiasg | I would like images better as well :) | 01:45 |
| @iglesiasg | I think it is more eye-candy | 01:45 |
| hushell | yeah, much better demo for ipython notebook :) | 01:46 |
| hushell | I will think about converting some of my work to here, but the feature part will be annoying | 01:47 |
| @iglesiasg | hushell, why? | 01:47 |
| @iglesiasg | why annotying I mean | 01:47 |
| hushell | Usually CRF features involves many types, color, texture, shape, position etc | 01:48 |
| hushell | That's quite a lot code, I have these in matlab | 01:48 |
| hushell | ah, you mean we can precompute these for demo? | 01:49 |
| @iglesiasg | hushell, yeah | 01:52 |
| @iglesiasg | hushell, or if you have them in matlab, we can just create bindings of your code for matlab | 01:52 |
| hushell | iglesiasg: can private functions have STL arguments? | 01:52 |
| @iglesiasg | it would be definitely better if your code can also be used from octave | 01:52 |
| @iglesiasg | hushell, yes, no problem | 01:53 |
| hushell | and private class members? | 01:53 |
| @iglesiasg | hushell, but it would be better if you can encapsulate those functions in another class so you don't need to include in the header the include of the STL class | 01:53 |
| hushell | will not be see outside | 01:53 |
| @iglesiasg | those functions and members then | 01:53 |
| @iglesiasg | it is what I have done with LMNN | 01:53 |
| @iglesiasg | I have LMNN and LMNNImpl with lot of Eigen and STL | 01:54 |
| hushell | good, I will have a look | 01:54 |
| @iglesiasg | nothing of those in LMNN.h, everything in LMNNImpl.h | 01:54 |
| @iglesiasg | you will see that I have done all the methods static in LMNNImpl.h | 01:54 |
| @iglesiasg | but that is because I was confused at first | 01:55 |
| @iglesiasg | I will probably change it to normal sooner or later | 01:55 |
| hushell | :) I see | 01:55 |
| hushell | iglesiasg: you put everything internal in LMNNImpl.h, good idea! | 01:57 |
| @iglesiasg | hushell, yes! It is something called opaque pointer or pimpl | 01:57 |
| @iglesiasg | http://en.wikipedia.org/wiki/Opaque_pointer | 01:57 |
| @iglesiasg | in QT they have made a science of that :D | 01:58 |
| hushell | iglesiasg: Thanks! Learned a lot! | 01:58 |
| hushell | Then, I need to refactor a bit my code :( | 01:59 |
| @iglesiasg | hushell, sonney2k liked the idea quite a bit | 01:59 |
| @iglesiasg | he says we can very much improve compilation time using that | 02:00 |
| hushell | I understand a bit now, so your opaque pointers are pointed to those static funcs in LMNNImpl.h | 02:06 |
| hushell | In the wiki's example, they have a class pointer defined in the main class, and your case with static funcs even better | 02:07 |
| @iglesiasg | hushell, yes | 02:07 |
| @iglesiasg | but I don't think it is actually even better... | 02:07 |
| @iglesiasg | it implies some limitations I think | 02:07 |
| @iglesiasg | having a class pointer is just fine I think | 02:08 |
| hushell | After read this, I have to say it's great idea for our development! | 02:08 |
| hushell | iglesiasg: better I mean, no extra class need to be defined in the main class | 02:08 |
| @iglesiasg | you don't even need to include the implementation class in the public one, just doing forward declaration of the class is enough (that was what I didn't know and first, and why I made everything static) | 02:08 |
| @iglesiasg | hushell, I am glad you liked it! | 02:09 |
| hushell | in this case, the hided information will never be found | 02:09 |
| hushell | I am leaving man. Good night! | 02:12 |
| @iglesiasg | see you! | 02:12 |
| hushell | Good night everybody | 02:12 |
| -!- hushell [~hushell@8-92.ptpg.oregonstate.edu] has quit [Quit: WeeChat 0.3.7] | 02:12 | |
| -!- pickle27 [~Kevin@d67-193-243-174.home3.cgocable.net] has joined #shogun | 02:51 | |
| -!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 02:56 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Quit: Ex-Chat] | 03:20 | |
| -!- pickle27 [~Kevin@d67-193-243-174.home3.cgocable.net] has quit [Quit: Leaving] | 03:23 | |
| shogun-buildbot | build #1456 of deb3 - modular_interfaces is complete: Failure [failed test python_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1456 blamelist: Soeren Sonnenburg <sonne@debian.org> | 04:11 |
| shogun-buildbot | build #477 of nightly_default is complete: Failure [failed test] Build details are at http://www.shogun-toolbox.org/buildbot/builders/nightly_default/builds/477 | 05:04 |
| -!- gsomix [~gsomix@178.45.89.48] has joined #shogun | 06:54 | |
| gsomix | hello | 06:54 |
| -!- zxtx [~zv@c-98-207-60-193.hsd1.ca.comcast.net] has joined #shogun | 09:19 | |
| -!- lambday [67157d4c@gateway/web/freenode/ip.103.21.125.76] has joined #shogun | 09:34 | |
| lambday | good morning | 09:38 |
| -!- van51 [~van51@ppp-94-66-76-131.home.otenet.gr] has joined #shogun | 10:24 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun | 10:25 | |
| -!- mode/#shogun [+o iglesiasg] by ChanServ | 10:25 | |
| @iglesiasg | morning | 10:25 |
| -!- naywhayare [~ryan@spoon.lugatgt.org] has quit [Ping timeout: 268 seconds] | 12:04 | |
| -!- naywhayare [~ryan@spoon.lugatgt.org] has joined #shogun | 12:05 | |
| -!- zxtx [~zv@c-98-207-60-193.hsd1.ca.comcast.net] has quit [Ping timeout: 256 seconds] | 12:17 | |
| -!- van51 [~van51@ppp-94-66-76-131.home.otenet.gr] has quit [Remote host closed the connection] | 12:30 | |
| -!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 12:33 | |
| shogun-notifier- | shogun: Fernando Iglesias :develop * 8f80504 / src/shogun/metric/LMNNImpl.cpp,src/shogun/metric/LMNNImpl.h: https://github.com/shogun-toolbox/shogun/commit/8f80504e00b326e9a5326054b1a510cd51fc4580 | 12:33 |
| shogun-notifier- | shogun: Faster exact computation of neighbors in LMNN | 12:33 |
| shogun-notifier- | shogun: Fernando Iglesias :develop * 6043056 / tests/unit/metric/LMNNImpl_unittest.cc: https://github.com/shogun-toolbox/shogun/commit/604305678e284c32c2d0516925cb112a13fa373f | 12:33 |
| shogun-notifier- | shogun: Fix missing UNREF in LMNNImpl unittest | 12:33 |
| shogun-notifier- | shogun: Fernando Iglesias :develop * f138cb7 / / (3 files): https://github.com/shogun-toolbox/shogun/commit/f138cb7afb0e0d24e45915c8861b6a57116cb9d0 | 12:33 |
| shogun-notifier- | shogun: Merge pull request #1352 from iglesias/feature/lmnn | 12:33 |
| shogun-notifier- | shogun: | 12:33 |
| shogun-notifier- | shogun: Feature/lmnn | 12:33 |
| -!- lisitsyn1 [~lisitsyn@5-28-21-18.clients.tlt.100megabit.ru] has joined #shogun | 12:34 | |
| -!- lisitsyn [~lisitsyn@5-28-17-136.clients.tlt.100megabit.ru] has quit [Read error: Operation timed out] | 12:35 | |
| shogun-buildbot | build #1147 of cyg1 - libshogun is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1147 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com> | 12:50 |
| -!- travis-ci [~travis-ci@ec2-107-20-60-156.compute-1.amazonaws.com] has joined #shogun | 12:58 | |
| travis-ci | [travis-ci] it's Fernando Iglesias's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/9803573 | 12:58 |
| -!- travis-ci [~travis-ci@ec2-107-20-60-156.compute-1.amazonaws.com] has left #shogun [] | 12:58 | |
| shogun-buildbot | build #1148 of cyg1 - libshogun is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1148 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com> | 13:02 |
| shogun-buildbot | build #1457 of deb3 - modular_interfaces is complete: Failure [failed test python_modular] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1457 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com> | 13:21 |
| shogun-buildbot | build #1149 of cyg1 - libshogun is complete: Failure [failed compile] Build details are at http://www.shogun-toolbox.org/buildbot/builders/cyg1%20-%20libshogun/builds/1149 blamelist: Fernando Iglesias <fernando.iglesiasg@gmail.com> | 13:36 |
| -!- lisitsyn1 [~lisitsyn@5-28-21-18.clients.tlt.100megabit.ru] has quit [Quit: Leaving.] | 14:03 | |
| shogun-buildbot | build #1458 of deb3 - modular_interfaces is complete: Success [build successful] Build details are at http://www.shogun-toolbox.org/buildbot/builders/deb3%20-%20modular_interfaces/builds/1458 | 14:06 |
| -!- lisitsyn [~lisitsyn@5-28-21-18.clients.tlt.100megabit.ru] has joined #shogun | 14:07 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 268 seconds] | 14:48 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun | 15:05 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Client Quit] | 15:05 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun | 15:05 | |
| -!- mode/#shogun [+o iglesiasg] by ChanServ | 15:05 | |
| -!- shogun-notifier- [~irker@7nn.de] has quit [Quit: transmission timeout] | 15:33 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 240 seconds] | 15:55 | |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has joined #shogun | 16:11 | |
| -!- mode/#shogun [+o iglesiasg] by ChanServ | 16:11 | |
| -!- lambday [67157d4c@gateway/web/freenode/ip.103.21.125.76] has quit [] | 18:24 | |
| @sonney2k | hey gsomix | 19:04 |
| gsomix | sonney2k, hey | 19:05 |
| gsomix | sonney2k, I thinking about interface of LibSVMFile. Look, at first column we have labels - so I added new method get_labels_vector. | 19:07 |
| gsomix | and other columns are sparse matrix/vector | 19:07 |
| gsomix | we have get_sparse_matrix in CFile for this case | 19:07 |
| gsomix | *I'm thinking | 19:08 |
| @sonney2k | gsomix, it is done exactly this way in CSparseFeatures | 19:10 |
| @sonney2k | gsomix, the label is optional though | 19:10 |
| @sonney2k | gsomix, btw I have a bug report for get_*_vector | 19:10 |
| gsomix | sonney2k, get_*_vector? | 19:11 |
| @sonney2k | gsomix, when you just have a single line with many numbers | 19:11 |
| @sonney2k | it should return all these numbers | 19:11 |
| @sonney2k | not only when you have one number per line | 19:11 |
| gsomix | sonney2k, can you show me example? | 19:12 |
| gsomix | I didn't get. | 19:12 |
| -!- vgorbati [~vgorbati@91.216.173.29] has joined #shogun | 19:24 | |
| gsomix | sonney2k, I just have tested csv reader with single line. | 19:39 |
| gsomix | works fine | 19:40 |
| -!- vgorbati [~vgorbati@91.216.173.29] has quit [Quit: vgorbati] | 19:41 | |
| @sonney2k | gsomix, err sorry one entry per line | 20:00 |
| @sonney2k | gsomix, ../data/label_train_regression.dat | 20:00 |
| @sonney2k | gsomix, labels_io_modular.py | 20:00 |
| gsomix | sonney2k, ah, ok. you want get all this number in one vector? | 20:02 |
| @sonney2k | gsomix, yes | 20:03 |
| @sonney2k | gsomix, so get*vector should work with both variants | 20:04 |
| gsomix | sonney2k, I'll add flag for this case, ok? | 20:04 |
| @sonney2k | gsomix, no | 20:05 |
| @sonney2k | gsomix, just make it work transparantly | 20:05 |
| gsomix | sonney2k, but I have LineReader, that read lines with now. | 20:06 |
| gsomix | *with '\n' at end | 20:06 |
| gsomix | should I make it more configurable? | 20:07 |
| gsomix | I mean LineReader that contains in CVSReader. | 20:07 |
| gsomix | * is contained | 20:09 |
| @sonney2k | gsomix, I don't get it | 20:09 |
| @sonney2k | gsomix, if you line reader reads a single line then fine | 20:09 |
| @sonney2k | gsomix, but for get_*vector you should read the whole file anyways and if it has > 1 line | 20:09 |
| @sonney2k | it should just read the other lines and build the vector from that | 20:10 |
| gsomix | whole file? O_O | 20:10 |
| gsomix | oh | 20:10 |
| @sonney2k | gsomix, I would even say that you could just use the matrix reading code for that | 20:11 |
| @sonney2k | and then ``convert'' the matrix to vector | 20:11 |
| @sonney2k | gsomix, errm what is surprising? | 20:11 |
| gsomix | I thought get_vector should, for example in csv, read one line and then should parse it. just get ONE vector stored in file (in model vector-features). | 20:13 |
| @sonney2k | gsomix, no these CFile based ops are at least currently parsing the full file | 20:14 |
| gsomix | ok, I'll do it in this way. | 20:14 |
| @sonney2k | gsomix, however the streaming* features will need it the way you are doing it now | 20:15 |
| gsomix | sonney2k, ok. so. now I should make it in "read-whole-file" way, right? | 20:19 |
| -!- zxtx [~zv@c-98-207-60-193.hsd1.ca.comcast.net] has joined #shogun | 20:21 | |
| gsomix | sonney2k, and sorry for this. I'm just thoughtless | 20:25 |
| @sonney2k | gsomix, well everything is ok just get vector not | 20:26 |
| @sonney2k | gsomix, just use get_*_matrix there and then if it has num_columns == 1 || num_rows==1 return sth reasonable | 20:26 |
| @sonney2k | by doing SGMatrix -> SGVector | 20:27 |
| lisitsyn | ahhhr sonney2k I forgot one thing | 20:32 |
| lisitsyn | about foulwall stuff | 20:32 |
| @sonney2k | lisitsyn, one thing? | 20:32 |
| @sonney2k | ahh | 20:32 |
| @sonney2k | what? | 20:32 |
| lisitsyn | sonney2k: that MIT-CBCL dataset has some strnage license | 20:32 |
| lisitsyn | http://cbcl.mit.edu/software-datasets/heisele/facerecognition-database.html | 20:32 |
| lisitsyn | 3. You may not distribute the downloaded images, or any portion thereof, to anyone else. | 20:32 |
| lisitsyn | sonney2k: ^ | 20:32 |
| @sonney2k | lisitsyn, ok so it has to be dropped then | 20:33 |
| lisitsyn | I guess so | 20:33 |
| lisitsyn | stupid license :( | 20:33 |
| @sonney2k | not a problem | 20:33 |
| @sonney2k | the other examples are sufficient | 20:34 |
| lisitsyn | this dataset is really really cool for DR | 20:34 |
| lisitsyn | it has real intrinsic dim of 2 | 20:34 |
| @sonney2k | lisitsyn, why not create an artificial one with the same properties? | 20:39 |
| lisitsyn | sonney2k: hah yes I wanted to | 20:40 |
| @sonney2k | lisitsyn, I mean you know take some 3d object rotate it or so | 20:40 |
| lisitsyn | yes | 20:40 |
| @sonney2k | actually I generated some data set like this for teaching. not that beautiful though | 20:46 |
| @sonney2k | IIRC was some 2d-plane rotated to be in some n-dim space | 20:46 |
| lisitsyn | well let foulwall do that then ;) | 20:46 |
| lisitsyn | it should be easy to script such a rotating in blender | 20:47 |
| @sonney2k | lisitsyn, I would rather move on and take the data we have | 20:48 |
| lisitsyn | yeah | 20:48 |
| lisitsyn | sure | 20:48 |
| lisitsyn | I am ok with it too | 20:48 |
| @sonney2k | lisitsyn, there are a couple of things that require manual work like hierarchical clustering | 20:49 |
| gsomix | sonney2k, btw did you got my email yeasterday? | 20:56 |
| gsomix | awful day. I was in city (parents was needed help) from 15:00 to 21:00. | 20:57 |
| gsomix | and 80% time we was in car. | 20:58 |
| gsomix | very, very hard traffic | 20:58 |
| gsomix | and many car accident at city | 20:58 |
| gsomix | *accidents | 20:58 |
| gsomix | the city has gone mad | 20:58 |
| -!- iglesiasg [~iglesias@s83-179-44-135.cust.tele2.se] has quit [Ping timeout: 264 seconds] | 21:11 | |
| @sonney2k | gsomix, yes I did no clue what this was about though | 21:12 |
| @sonney2k | but I will have the same tomorrow getting back home | 21:12 |
| gsomix | sonney2k, huh, ok. | 21:13 |
| -!- gsomix_ [~gsomix@109.169.187.164] has joined #shogun | 21:20 | |
| -!- gsomix [~gsomix@178.45.89.48] has quit [Ping timeout: 256 seconds] | 21:24 | |
| -!- pickle27 [~Kevin@67.193.243.174] has joined #shogun | 22:14 | |
| lisitsyn | pickle27: hey | 22:16 |
| pickle27 | hey! | 22:17 |
| lisitsyn | pickle27: how is it going? | 22:17 |
| pickle27 | lisitsyn: good! you? | 22:17 |
| lisitsyn | pickle27: quite good too, last week was quite intense | 22:18 |
| lisitsyn | I moved to another place and a lot of things at my job | 22:18 |
| pickle27 | lots of work? | 22:18 |
| pickle27 | ahh | 22:18 |
| lisitsyn | pickle27: yeah we had kind of deadline to have a showcase | 22:19 |
| lisitsyn | it failed though haha | 22:19 |
| pickle27 | oh no! | 22:19 |
| lisitsyn | but that's not an issue of mine, we have other people to sell things | 22:19 |
| pickle27 | right cool | 22:21 |
| pickle27 | I defended my thesis on Thurs! | 22:21 |
| lisitsyn | oh | 22:21 |
| lisitsyn | :D | 22:21 |
| lisitsyn | I missed that | 22:21 |
| lisitsyn | so how it was? | 22:21 |
| pickle27 | lisitsyn: it went well! | 22:21 |
| lisitsyn | so you are Msc now? | 22:21 |
| pickle27 | MASc | 22:22 |
| lisitsyn | what does it mean? | 22:22 |
| pickle27 | I'm under comp eng so its a masters of applied science but essentially yes | 22:22 |
| lisitsyn | ahh | 22:22 |
| lisitsyn | no clue about that :D | 22:22 |
| pickle27 | lisitsyn: everything is going well though. I should have a PR for my audio example as an ipython notebook next week | 22:23 |
| lisitsyn | pickle27: good! | 22:24 |
| lisitsyn | pickle27: when do you start your job? | 22:24 |
| pickle27 | lisitsyn: not until November | 22:24 |
| lisitsyn | I see! | 22:24 |
| pickle27 | lisitsyn: hows cmake coming along? | 22:31 |
| lisitsyn | pickle27: shame for me I am not that helpful for wiking but he is doing really well | 22:32 |
| lisitsyn | better and better :) | 22:32 |
| pickle27 | lisitsyn: im excited for it | 22:32 |
| lisitsyn | yeah me too | 22:32 |
| @sonney2k | lisitsyn, btw did you have a chance to look at that bug | 23:38 |
| lisitsyn | sonney2k: right now | 23:38 |
| @sonney2k | thx | 23:38 |
| lisitsyn | sonney2k: Heiko wins I will add some tests ;) | 23:39 |
| @sonney2k | lisitsyn, heh yeah and wiking - they were very pushy wrt | 23:42 |
| @sonney2k | I even started writing tests my own :) | 23:42 |
| lisitsyn | sonney2k: yeah they make the world go round | 23:42 |
| -!- shogun-notifier- [~irker@7nn.de] has joined #shogun | 23:58 | |
| shogun-notifier- | shogun: Soeren Sonnenburg :develop * c1bbb34 / examples/undocumented/python_modular/features_dense_io_modular.py: https://github.com/shogun-toolbox/shogun/commit/c1bbb3468650aa04cb9130d7afae6fa251cf984b | 23:58 |
| shogun-notifier- | shogun: fix function name in example | 23:58 |
| --- Log closed Sun Aug 04 00:00:00 2013 | ||
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!