--- Log opened Fri Apr 12 00:00:27 2013 | ||
-!- iglesiasg [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 00:06 | |
-!- medeeiip_ [~medeeiip@116.193.143.188] has quit [Quit: Leaving] | 00:25 | |
-!- mgahlot|away [~quassel@120.59.78.2] has joined #shogun | 01:04 | |
-!- mgahlot|away is now known as mgahlot | 01:08 | |
-!- mgahlot|away [~quassel@59.177.128.226] has joined #shogun | 01:14 | |
-!- mgahlot [~quassel@120.59.78.2] has quit [Ping timeout: 245 seconds] | 01:17 | |
-!- mgahlot|away is now known as mgahlot | 01:18 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has joined #shogun | 02:22 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has quit [Remote host closed the connection] | 02:25 | |
-!- mgahlot is now known as mgahlot|away | 03:38 | |
-!- iglesiasg [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 04:06 | |
-!- mgahlot|away [~quassel@59.177.128.226] has quit [Ping timeout: 276 seconds] | 04:18 | |
-!- gsomix [~Miranda@185.6.239.119] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org] | 05:58 | |
-!- Yanglittle [deb20afe@gateway/web/freenode/ip.222.178.10.254] has joined #shogun | 06:03 | |
Yanglittle | Hey, how can I get the info of methods in detail except help , eg. shogun.Features.CombinedFeatures.append_features.obj(), | 06:05 |
---|---|---|
Yanglittle | I'm reading the examples, and not know the meaning of the methods clearly. | 06:07 |
-!- foulwall [~foulwall@2001:da8:215:c252:f127:a5c4:7b9a:d2ed] has joined #shogun | 07:38 | |
-!- foulwall [~foulwall@2001:da8:215:c252:f127:a5c4:7b9a:d2ed] has quit [Ping timeout: 245 seconds] | 07:43 | |
-!- foulwall [~foulwall@2001:da8:215:c252:6036:ed:476a:c97f] has joined #shogun | 07:46 | |
foulwall | anybody there? | 07:46 |
foulwall | sonney2k: Arrange my today's work, pls | 07:47 |
foulwall | ...caught a bad cold... but no problem I'll try my best | 07:51 |
-!- foulwall [~foulwall@2001:da8:215:c252:6036:ed:476a:c97f] has left #shogun [] | 07:51 | |
-!- lambday [0e8b6149@gateway/web/freenode/ip.14.139.97.73] has joined #shogun | 08:00 | |
-!- sploving [~sploving@124.16.141.74] has joined #shogun | 08:00 | |
-!- foulwall [~foulwall@2001:da8:215:6819:e1ec:d3c4:f66c:c180] has joined #shogun | 08:32 | |
@sonney2k | foulwall, then better recover before you continue | 08:51 |
-!- foulwall [~foulwall@2001:da8:215:6819:e1ec:d3c4:f66c:c180] has quit [Remote host closed the connection] | 08:52 | |
@sonney2k | Yanglittle, that is the part where shogun needs improvement. we have lots of methods / functions but documentation /tutorial is not as good | 08:53 |
-!- foulwall [~foulwall@2001:da8:215:6819:e1ec:d3c4:f66c:c180] has joined #shogun | 08:53 | |
@sonney2k | foulwall, once you feel better we should try to get your regression demo into the shogun homepage | 08:54 |
foulwall | sonney2k: maybe tomorrow? | 08:55 |
@sonney2k | foulwall, ok - please check out https://github.com/shogun-toolbox/shogun-web tomorrow - we could then under some link say /demos/regression put in the examples | 08:59 |
foulwall | sonney2k: gotcha | 09:00 |
@sonney2k | foulwall, but take care health is more important! | 09:00 |
foulwall | sonney2k: thanks, i will. | 09:00 |
-!- sploving [~sploving@124.16.141.74] has left #shogun [] | 09:03 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:06 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:07 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:12 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:13 | |
-!- lisitsyn [~blackburn@185.6.239.119] has quit [Quit: Leaving.] | 09:19 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:19 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:21 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:30 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:31 | |
-!- erlenda [~erlenda@fw-oslo.intelcom.no] has joined #shogun | 09:33 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:40 | |
-!- foulwall [~foulwall@2001:da8:215:6819:e1ec:d3c4:f66c:c180] has quit [Remote host closed the connection] | 09:41 | |
-!- sumit_ [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 09:41 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:41 | |
-!- medeeiip [~medeeiip@103.2.133.140] has joined #shogun | 09:42 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:47 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:49 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:52 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:53 | |
-!- gzhd79 [~zeller@embln.embl.de] has joined #shogun | 09:54 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 09:56 | |
-!- sumit_ [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 09:56 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 09:57 | |
-!- gzhd79 [~zeller@embln.embl.de] has left #shogun [] | 10:00 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:01 | |
-!- blackburn [~lisitsin@mxs.kg.ru] has joined #shogun | 10:02 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:03 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:07 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:08 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:14 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:15 | |
-!- iglesiasg [~nando@n153-p188.kthopen.kth.se] has joined #shogun | 10:19 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:19 | |
iglesiasg | blackburn is back? | 10:20 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:21 | |
blackburn | iglesiasg: oops | 10:22 |
-!- blackburn is now known as lisitsyn | 10:22 | |
-!- mode/#shogun [+o lisitsyn] by ChanServ | 10:22 | |
@lisitsyn | I should get used to that too :D | 10:23 |
iglesiasg | hehe | 10:23 |
iglesiasg | you have superpowers now | 10:23 |
-!- andrei_iusan [51c48aec@gateway/web/freenode/ip.81.196.138.236] has joined #shogun | 10:23 | |
-!- medeeiip [~medeeiip@103.2.133.140] has quit [Quit: Leaving] | 10:24 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 10:26 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun | 10:27 | |
lisitsyn | blabla | 10:27 |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Client Quit] | 10:27 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has joined #shogun | 10:27 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:27 | |
sonne|work | iglesiasg: what happened to n4nd0? | 10:28 |
iglesiasg | I like this nick better | 10:29 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:29 | |
iglesiasg | because of my github identity | 10:29 |
iglesiasg | I think it looks more serious :) | 10:29 |
-!- mode/#shogun [+o lisitsyn] by ChanServ | 10:29 | |
sonne|work | iglesiasg: so you registered it? | 10:29 |
iglesiasg | sonne|work: yes | 10:29 |
sonne|work | lisitsyn: btw you will receive a t-shirt | 10:30 |
sonne|work | but a mentor t-shirt | 10:30 |
@lisitsyn | sonne|work: no, they do not ship to nigeria | 10:31 |
sonne|work | and when you visit the summit you get another one | 10:31 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:33 | |
iglesiasg | lisitsyn: maybe you could have it shipped to Germany since you are going there for the workshop | 10:34 |
@lisitsyn | iglesiasg: hah well I do not care that much about it | 10:34 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:34 | |
@lisitsyn | iglesiasg: anyway may be it is true only for students | 10:35 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:38 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:39 | |
iglesiasg | let me ask you a basic ML question regarding the classical SVM formulation for classification | 10:39 |
sonne|work | lisitsyn: ship it to my address and I give it to you when you are here :D | 10:40 |
lambday | hi lisitsyn | 10:40 |
iglesiasg | so the objective looks like: \min_{w, \Xi} ||w||^2 + C \sum \Xi_i | 10:40 |
iglesiasg | the regularization part, ||w||^2 | 10:41 |
iglesiasg | I think of this part as that is the one that actually takes care of margin maximization; it can be seen equivalent to \max M where M is the margin and fixing the norm of w to 1/M | 10:42 |
@lisitsyn | sonne|work: makes sense - I should ask carol | 10:42 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:42 | |
@lisitsyn | lambday: hi | 10:42 |
@lisitsyn | iglesiasg: yes regularization is margin | 10:42 |
iglesiasg | but I remember once that Georg told me that it was also there because it can be proven that an arbitraly high norm of w can fit perfectly to data | 10:43 |
iglesiasg | so it is there actually to avoid overfitting | 10:43 |
@lisitsyn | iglesiasg: yes bigger margin - less overfitting | 10:43 |
iglesiasg | I guess it is another point of view of the same concept, however I think there is something that I miss | 10:43 |
@lisitsyn | iglesiasg: there are a few views on that | 10:43 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:43 | |
lambday | I've been trying to use SGVector constructor (data, vlen).. when I take a pointer and use new, its all fine but when I don't I'm getting a weird error... | 10:43 |
@lisitsyn | iglesiasg: another thing is | 10:43 |
iglesiasg | lisitsyn: but why arbitrarily high norm of w can perfectly fit data? | 10:44 |
@lisitsyn | iglesiasg: in geometrical or? | 10:44 |
@lisitsyn | lambday: what is the error? | 10:44 |
iglesiasg | lisitsyn: I don't know, in any way it can be understood - not necessarily geometrically if it cannot be seen like that | 10:44 |
@lisitsyn | iglesiasg: well if you don't care about margin (regularization) | 10:45 |
@lisitsyn | i.e. your C is low | 10:45 |
@lisitsyn | you minimize the error on the training set | 10:45 |
@lisitsyn | w/o any restrictions | 10:45 |
@lisitsyn | so you fit perfectly or so | 10:45 |
iglesiasg | lisitsyn: mmm | 10:45 |
lambday | corrupted double-linked list: 0x00000000017f0590 *** | 10:46 |
lambday | and then it prints the memory map | 10:46 |
@lisitsyn | iglesiasg: speaking geometrical margin has some other things | 10:46 |
@lisitsyn | iglesiasg: IIRC VC dimension has something to do with margin of a linear classifier | 10:46 |
@lisitsyn | iglesiasg: so it stands for complexity | 10:47 |
@lisitsyn | iglesiasg: the other nice view on that is | 10:47 |
@lisitsyn | imagine you have w = [10,0.1,0.1] | 10:47 |
iglesiasg | lisitsyn: ok | 10:47 |
@lisitsyn | and w = [2,2,2] | 10:47 |
@lisitsyn | iglesiasg: latter is better because the distribution of coefficients is more smooth or so | 10:48 |
@lisitsyn | iglesiasg: in the first case you get feature 1 *very important* | 10:48 |
iglesiasg | I see | 10:48 |
@lisitsyn | so regularization denies one to 'overfit' like setting big coefficients to some features | 10:48 |
@lisitsyn | iglesiasg: w/o regularization it could get like 1003402340, -32132132, 54545 | 10:49 |
@lisitsyn | but it would fit the data | 10:49 |
iglesiasg | that is why I see not so clearly | 10:49 |
@lisitsyn | with no real induction | 10:49 |
iglesiasg | why bigger values fit better the data? | 10:49 |
@lisitsyn | iglesiasg: that's not a rule it could happen | 10:49 |
iglesiasg | aham I understand | 10:50 |
@lisitsyn | iglesiasg: there are so many views on that | 10:50 |
iglesiasg | but say two instances of w | 10:51 |
iglesiasg | both of them of similar smoothness | 10:51 |
@lisitsyn | iglesiasg: if I was a physician I'd say something about fredholm equation :D | 10:51 |
iglesiasg | and both of them fitting the data more or less in the same way - so with \xi_i values sort of similar | 10:51 |
@lisitsyn | iglesiasg: it is pretty illposed problem so would not expect that | 10:51 |
iglesiasg | if one of the vectors has high values and the other one low values | 10:52 |
@lisitsyn | :) | 10:52 |
iglesiasg | why the one with low values should be preferred? | 10:52 |
@lisitsyn | iglesiasg: why vectors? | 10:52 |
@lisitsyn | are you talking about dual? | 10:52 |
iglesiasg | I am calling w vector | 10:52 |
@lisitsyn | ahh | 10:52 |
@lisitsyn | so you are asking why ||w|| -> min is preferred? | 10:52 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 10:53 | |
iglesiasg | well, because of definition ||w|| smaller implies larger margin I guess | 10:53 |
@lisitsyn | 1) it stands for margin (2/||w|| is the margin) | 10:53 |
@lisitsyn | 2) it stands for complexity (implied from margin) | 10:53 |
@lisitsyn | iglesiasg: it is not just implies - margin is exactly 2/||w| | 10:54 |
iglesiasg | agree | 10:54 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 10:54 | |
@lisitsyn | in VC theory they kind of proved that margin is related to complexity | 10:54 |
@lisitsyn | iglesiasg: I like the complexity view - it unifies that framework with say decision trees | 10:54 |
@lisitsyn | in decision trees you have no margin but you can derive something with MDL | 10:55 |
@lisitsyn | minimum description length | 10:55 |
iglesiasg | aham | 10:55 |
@lisitsyn | like the smaller your tree is the better it is | 10:55 |
iglesiasg | very nice theory indeed | 10:56 |
@lisitsyn | I am wondering if some day some guy with proper brain would come with something better | 10:56 |
iglesiasg | that's why the world needs sonne|work to come back to research :) | 10:58 |
@lisitsyn | I am afraid there is no such person yet at all | 10:59 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 11:00 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 11:02 | |
iglesiasg | lisitsyn: http://abstrusegoose.com/502 probably we will never know | 11:06 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 11:06 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 11:08 | |
andrei_iusan | hello, I'm a student at CS, interested in your projects, especially in BSS | 11:08 |
iglesiasg | hi andrei_iusan, that is good | 11:08 |
andrei_iusan | I have studied linear algebra and numerical methods | 11:08 |
andrei_iusan | I think those are the most relevant courses | 11:09 |
-!- foo__ [2eda6d58@gateway/web/freenode/ip.46.218.109.88] has joined #shogun | 11:09 | |
andrei_iusan | I worked with Matlab/Octave | 11:09 |
foo__ | Hi all | 11:09 |
foo__ | I am wondering if Platt's method for probabilistic output svm is implemented into shogun ? I remember a discussion on this in the forum... | 11:10 |
@lisitsyn | foo__: yes | 11:11 |
@lisitsyn | foo__: scores_to_probabilities method of BinaryLabels | 11:12 |
andrei_iusan | I haven't studied machine learning algorithms, but I would like too, is it reallistic that I could study and learn the algorithms and technics, or I should have more knowledge for this project? | 11:13 |
iglesiasg | andrei_iusan: I think it is possible to learn them | 11:14 |
@lisitsyn | andrei_iusan: yes sure, you don't have to be a phd in machine learning :) | 11:14 |
andrei_iusan | I know matlab and played with matrices, factorizations, Jacobi, Givens, eigen-values, eigen-vectors, diagonalization, etc | 11:15 |
-!- foo__ [2eda6d58@gateway/web/freenode/ip.46.218.109.88] has quit [Ping timeout: 245 seconds] | 11:15 | |
@lisitsyn | andrei_iusan: that's cool for sure | 11:15 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 11:15 | |
@lisitsyn | andrei_iusan: you might need to consider contributing with something small though | 11:15 |
@lisitsyn | andrei_iusan: are you subscribed to the mailing list? | 11:16 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 11:16 | |
andrei_iusan | no, not yet, I will subscribe now | 11:16 |
andrei_iusan | ok, I have subscribed | 11:19 |
-!- foo_ [2eda6d58@gateway/web/freenode/ip.46.218.109.88] has joined #shogun | 11:19 | |
@lisitsyn | andrei_iusan: http://blog.gmane.org/gmane.comp.ai.machine-learning.shogun check archives (last week or so) - there is some information | 11:20 |
andrei_iusan | ok | 11:20 |
foo_ | Is probabilistic output for svm is available in shogun ? | 11:21 |
@lisitsyn | foo_: scores_to_probabilities method of BinaryLabels | 11:21 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 11:23 | |
@lisitsyn | iglesiasg: ah btw that comma madness works | 11:24 |
iglesiasg | lisitsyn: cool | 11:24 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 11:24 | |
@lisitsyn | iglesiasg: will commit tonight if I manage to fix the last issue with default values | 11:24 |
iglesiasg | I will have a look once it is up | 11:25 |
-!- foo_ [2eda6d58@gateway/web/freenode/ip.46.218.109.88] has quit [Ping timeout: 245 seconds] | 11:25 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 11:34 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 11:35 | |
-!- hoijui [~hoijui@dslb-088-074-124-102.pools.arcor-ip.net] has joined #shogun | 11:41 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has joined #shogun | 11:50 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 11:52 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 11:53 | |
andrei_iusan | ok, so what do I do next? :D | 11:58 |
andrei_iusan | check the issues on github? | 11:58 |
@lisitsyn | andrei_iusan: yes | 11:58 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 11:58 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 12:00 | |
andrei_iusan | ok | 12:00 |
-!- heiko1 [~heiko@nat-166-51.internal.eduroam.ucl.ac.uk] has joined #shogun | 12:04 | |
-!- andrei_iusan [51c48aec@gateway/web/freenode/ip.81.196.138.236] has quit [Quit: Page closed] | 12:04 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has quit [Remote host closed the connection] | 12:06 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 12:12 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 12:13 | |
lambday | hi heiko1 | 12:15 |
heiko1 | lambday: hi! | 12:15 |
lambday | heiko1: in the sampling gaussian unit test for dense... shall I again use fixed cov matrix | 12:15 |
heiko1 | lambday: what do you mean by fixed? | 12:16 |
lambday | heiko1: not random... fixed values | 12:16 |
heiko1 | lambday: I see, no random is better there | 12:16 |
heiko1 | just compare empirical covariance with the original one | 12:16 |
lambday | that's what I was thinking :) | 12:16 |
lambday | yes.. norm.. | 12:17 |
lambday | one more thing.. as of now, i'm testing both empirical mean and cov in the same test | 12:17 |
heiko1 | lambday: yes, just frobenius norm or whatever | 12:17 |
heiko1 | lambday: thats fine | 12:17 |
lambday | heiko1: okay... | 12:18 |
lambday | heiko1: thanks man :) | 12:18 |
heiko1 | lambday: one more thing | 12:18 |
lambday | heiko1: yes | 12:18 |
heiko1 | could you add a boolean flag to the method | 12:18 |
lambday | for? | 12:18 |
heiko1 | called "precision_matrix" whose default is false | 12:19 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 12:19 | |
heiko1 | and if it is true, the matrix is regarded as the inverse of the covariance matrix | 12:19 |
heiko1 | so for a matrix C | 12:19 |
heiko1 | one sometimes wants to sample N(0, C) and sometimes N(0, C^(-1)) | 12:19 |
lambday | heiko1: okay... | 12:19 |
heiko1 | lambday: I can send you matlab code how the sampling works if needed | 12:20 |
heiko1 | basically just solve L\randn(...) system | 12:20 |
heiko1 | you can use eigen for that | 12:20 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 12:20 | |
-!- shubham_ [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #shogun | 12:20 | |
lambday | heiko1: if precision_matrix is false, it would use C, otherwise it would use C_inv, in that case, it would use the cholesky factor of C_inv, right? | 12:21 |
heiko1 | lambday: exactly, however, do not invert the matrix (since numerical problems) | 12:21 |
heiko1 | just invert the cholesky instead | 12:21 |
lambday | heiko1: okay.. | 12:22 |
heiko1 | so in anyway, you have to compute C=L*L' | 12:22 |
heiko1 | and then either you do L*randn(...) or (L^T )^(-1)*randn( ...) | 12:22 |
lambday | ummm.. | 12:23 |
lambday | yes | 12:23 |
heiko1 | lambday: https://gist.github.com/karlnapf/5371082 | 12:23 |
heiko1 | this is what I use in Matlab | 12:24 |
heiko1 | in fact, it would be great if you could also offer these standard arguments | 12:24 |
lambday | heiko1: thanks... in the current code, I took a sample from N(0,I) of size Nxdim instead (ease of use with matrix_mean and matrix_cov method in Statistics.h).. | 12:25 |
heiko1 | z is the standard Gaussian vector (if given, do not generate one) | 12:25 |
heiko1 | inds is a permutation that you want to use | 12:25 |
heiko1 | matrix_mean? matrix_cov? | 12:25 |
lambday | heiko1: matrix_mean for calculating empirical mean.. | 12:26 |
heiko1 | lambday: you mean in the test? | 12:26 |
lambday | heiko1: yes .. | 12:26 |
heiko1 | yes, thats fine fine | 12:26 |
lambday | so, I used upper cholesky factor instead | 12:27 |
heiko1 | oh wait, but please test on another matrix than I | 12:27 |
heiko1 | identity | 12:27 |
heiko1 | and non-zero mean | 12:27 |
heiko1 | lower/upper doesnt matter | 12:27 |
heiko1 | as long as the samples are correct :) | 12:27 |
lambday | heiko1: hmm.. | 12:27 |
lambday | heiko1: for the normal case (precision_matrix = false), current code is working fine.. the empirical and the actual mean and cov are quite close.. | 12:28 |
heiko1 | lambday: also for non-zero, non-identity case? | 12:30 |
lambday | heiko1: yes.. :) | 12:30 |
heiko1 | good! | 12:30 |
heiko1 | if you want, we can already merge this - and then continue with the rest | 12:30 |
lambday | heiko1: few things left for the dense case.. | 12:30 |
heiko1 | did you follow the instructions of Sergey's mail? | 12:30 |
heiko1 | lambday: yes? | 12:31 |
lambday | heiko1: for the gitflow? | 12:31 |
heiko1 | lambday: yes | 12:31 |
lambday | heiko1: watched the video link you gave.. didn't try though... | 12:31 |
lambday | :( | 12:32 |
heiko1 | dont worry | 12:32 |
lambday | heiko1: I did all the changes locally since then | 12:32 |
heiko1 | ok | 12:32 |
heiko1 | so first of all, create a backup :) | 12:33 |
lambday | heiko1: okay.. | 12:33 |
heiko1 | then run git flow init | 12:33 |
heiko1 | it will ask you a few things | 12:33 |
heiko1 | select develop and master for release and production | 12:34 |
heiko1 | and default values for the rest | 12:35 |
heiko1 | develop might not be in there | 12:35 |
heiko1 | oh and shogun_ as version tag | 12:35 |
lambday | heiko1: shall I commit first before doing this? I am using a temporary branch locally for this.. | 12:36 |
heiko1 | no, just make a backup | 12:36 |
heiko1 | you can replace the files later and commit from the new structure | 12:36 |
heiko1 | but make sure, things are not deleted by accident :) | 12:36 |
lambday | heiko1: alright... | 12:36 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 12:37 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 12:38 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 12:42 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 12:44 | |
heiko1 | lambday: and? | 12:44 |
lambday | heiko1: I am using fedora and can't get gitflow using yum install... its not in their repo yet | 12:44 |
lambday | so, building from the source :( | 12:44 |
heiko1 | lambday: uh, I see, should be fine doing this | 12:45 |
iglesiasg | sonne|work: so it seems that it is Spain vs Germany in the champions league :) | 12:45 |
lambday | heiko1: argh.. I ran make install from gitflow and it asks "Run 'git submodule init && git submodule update' first."... when I do that, it says connection timed out :'( | 12:48 |
heiko1 | lambday: mmmh | 12:51 |
heiko1 | do you have checked out the submodules? | 12:51 |
heiko1 | it is the data folder | 12:51 |
-!- iglesiasg [~nando@n153-p188.kthopen.kth.se] has quit [Read error: Operation timed out] | 12:52 | |
heiko1 | oh | 12:52 |
heiko1 | in fact, that might be the gitflow git | 12:52 |
lambday | heiko1: didn't get it :( | 12:53 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 12:53 | |
heiko1 | this git submodule is probably for the gitflow | 12:54 |
heiko1 | Ill download it and try to install to see what happens | 12:54 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 12:55 | |
heiko1 | lambday: the commands worked here | 12:55 |
heiko1 | just do "git submodule init && git submodule update" inside the gitflow folder | 12:55 |
lambday | heiko1: ya.. it shows cloning into 'shFlags' but then connection timed out.. | 12:57 |
heiko1 | lambday: maybe re-checkout retry, its working here | 12:58 |
lambday | heiko1: okay.. i'm trying | 12:58 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 13:01 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 13:03 | |
heiko1 | lambday: those admin things are always annoying since they consume so much time :( | 13:04 |
lambday | heiko1: yes... :( | 13:05 |
lambday | I guess its proxy error... I have https proxy set but it uses git:// .. | 13:05 |
heiko1 | lambday: yes certainly | 13:06 |
heiko1 | works all fine here | 13:06 |
heiko1 | I think you can tell git the proxy | 13:06 |
heiko1 | saw it somewhere | 13:06 |
lambday | hmm.. i'm trying.. | 13:08 |
-!- iglesiasg [~nando@n153-p188.kthopen.kth.se] has joined #shogun | 13:10 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 13:13 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 13:14 | |
-!- Yanglittle [deb20afe@gateway/web/freenode/ip.222.178.10.254] has quit [Quit: Page closed] | 13:28 | |
lambday | heiko1: finally gitflow installed... phew! | 13:48 |
lambday | heiko1: Branch name for production releases: [master] ... I'm confused.. shall i put develop here or master? and yes you were right develop isn't here.. just showing master and another branch i've created locally | 13:53 |
heiko1 | if develop is not there | 13:53 |
heiko1 | create a branch | 13:53 |
heiko1 | and then pull from upstream | 13:53 |
heiko1 | lisitsyn: right? | 13:53 |
heiko1 | then it is there | 13:54 |
@lisitsyn | heiko1: yes iirc | 13:54 |
@lisitsyn | it is a fork? | 13:54 |
lambday | it is a fork | 13:55 |
heiko1 | lisitsyn: could you guide lambday thourgh the process of creating a PR? I am going out for lunch now | 13:55 |
@lisitsyn | heiko1: good timing! | 13:56 |
@lisitsyn | ;) | 13:56 |
heiko1 | lambday: ask lisitsyn, I will be back in 1 hr | 13:56 |
heiko1 | lisitsyn: well we are trying for some time now :) | 13:56 |
lambday | heiko1: no problem.. ciao :) | 13:56 |
@lisitsyn | heiko1: 1 hr! | 13:56 |
@lisitsyn | :D | 13:56 |
heiko1 | lisitsyn: + coffee :D | 13:56 |
@lisitsyn | it must be an elephant | 13:57 |
@lisitsyn | om nom | 13:57 |
@lisitsyn | lambday: okay let me reproduce your issue here | 13:57 |
lambday | :D | 13:57 |
@lisitsyn | and then I'll let you know how to proceed | 13:57 |
lambday | hmm | 13:57 |
@lisitsyn | lambday: what does 'git remote -v' outputs? | 13:59 |
lambday | originhttps://github.com/lambday/shogun (fetch) originhttps://github.com/lambday/shogun (push) upstreamhttps://github.com/shogun-toolbox/shogun.git (fetch) upstreamhttps://github.com/shogun-toolbox/shogun.git (push) | 14:00 |
sonne|work | heiko1: could you please register with nickserv? | 14:00 |
heiko1 | sonne|work: yes will do, | 14:00 |
@lisitsyn | lambday: please do 'git fetch upstream' | 14:01 |
@lisitsyn | and develop should appear | 14:01 |
sonne|work | heiko1: ohh and please be a bit more positive when you reply to students on the mailinglist - your replies are usually rough german style :D | 14:01 |
lambday | lisitsyn: okay | 14:01 |
heiko1 | sonne|work: haha, sure! :) | 14:01 |
lambday | lisitsyn: done | 14:02 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Read error: Connection reset by peer] | 14:02 | |
-!- mgahlot [~quassel@59.177.129.223] has joined #shogun | 14:02 | |
lambday | lisitsyn: yes develop is there | 14:02 |
@lisitsyn | lambday: production releases is master | 14:02 |
-!- mgahlot is now known as mgahlot|away | 14:02 | |
@lisitsyn | lambday: next version is develop | 14:02 |
@lisitsyn | others are default | 14:02 |
lambday | lisitsyn: okay... let me try | 14:03 |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 14:03 | |
lambday | lisitsyn: but when I do git flow init, it still shows master and <my-local-branch>.. | 14:04 |
@lisitsyn | no develop? | 14:04 |
lambday | nope.. | 14:04 |
@lisitsyn | lambday: git checkout develop | 14:04 |
@lisitsyn | and try again | 14:04 |
lambday | okay | 14:05 |
lambday | yes it shows now... | 14:05 |
lambday | Version tag prefix? [] ====> shogun_ ?? | 14:06 |
@lisitsyn | lambday: yes shogun_ but it doesn't really matter | 14:07 |
@lisitsyn | as you are not going to release yet :) | 14:07 |
lambday | okay | 14:07 |
lambday | lisitsyn: so, now if I want to commit, first i need to do git flow "sample_gaussian_dense" ... and then commit? | 14:10 |
lambday | i mean, git flow start "sample_gaussian_dense" | 14:11 |
lambday | then commits and then git flow finish "sample_gaussian_dense"... and then push.. | 14:12 |
@lisitsyn | lambday: git flow feature start sample_gaussian_dense | 14:15 |
@lisitsyn | some commits | 14:15 |
@lisitsyn | then git flow feature finish sample_gaussian_dense | 14:15 |
lambday | lisitsyn: alright.. | 14:16 |
lambday | and then push with git push origin develop..? | 14:16 |
@lisitsyn | lambday: yes | 14:16 |
lambday | lisitsyn: then normal PR.. gotcha | 14:16 |
@lisitsyn | yes | 14:17 |
lambday | lisitsyn: thanks man :) | 14:17 |
-!- medeeiip [~medeeiip@103.2.133.140] has joined #shogun | 14:50 | |
-!- shubham_ [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Ping timeout: 245 seconds] | 14:51 | |
-!- lambday [0e8b6149@gateway/web/freenode/ip.14.139.97.73] has quit [Ping timeout: 245 seconds] | 14:56 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 15:13 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 15:14 | |
-!- iglesiasg [~nando@n153-p188.kthopen.kth.se] has quit [Ping timeout: 240 seconds] | 15:17 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 15:19 | |
-!- mgahlot [~quassel@120.56.201.43] has joined #shogun | 15:28 | |
-!- mgahlot|away [~quassel@59.177.129.223] has quit [Ping timeout: 252 seconds] | 15:31 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 15:41 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 15:42 | |
-!- iglesiasg [~nando@n153-p188.kthopen.kth.se] has joined #shogun | 15:43 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 15:47 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 15:48 | |
-!- mgahlot is now known as mgahlot|away | 15:50 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 15:57 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 15:57 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:00 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:01 | |
-!- mgahlot|away [~quassel@120.56.201.43] has quit [Ping timeout: 260 seconds] | 16:02 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:04 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:05 | |
-!- iglesiasg [~nando@n153-p188.kthopen.kth.se] has quit [Quit: leaving] | 16:07 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:10 | |
-!- medeeiip__ [~medeeiip@103.2.133.140] has joined #shogun | 16:11 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:11 | |
-!- medeeiip [~medeeiip@103.2.133.140] has quit [Ping timeout: 258 seconds] | 16:13 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:14 | |
-!- lisitsyn [~lisitsin@mxs.kg.ru] has quit [Quit: Leaving.] | 16:15 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:15 | |
-!- erlenda [~erlenda@fw-oslo.intelcom.no] has quit [Quit: Leaving] | 16:17 | |
-!- mgahlot [~quassel@120.56.207.251] has joined #shogun | 16:17 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:20 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:21 | |
-!- mgahlot|away [~quassel@120.56.198.189] has joined #shogun | 16:22 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:25 | |
-!- mgahlot [~quassel@120.56.207.251] has quit [Ping timeout: 252 seconds] | 16:25 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:26 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:29 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:31 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:34 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:36 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Read error: Connection reset by peer] | 16:39 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:40 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 16:40 | |
-!- abinashpanda [012612ec@gateway/web/freenode/ip.1.38.18.236] has joined #shogun | 16:42 | |
-!- lambday [0e8b6149@gateway/web/freenode/ip.14.139.97.73] has joined #shogun | 16:42 | |
lambday | heiko1: hello | 16:42 |
heiko1 | lambday: hi! | 16:43 |
heiko1 | how is it going? | 16:43 |
lambday | https://github.com/lambday/shogun/blob/develop/src/shogun/mathematics/Statistics.cpp#L2061 | 16:43 |
lambday | https://github.com/lambday/shogun/blob/develop/tests/unit/mathematics/Statistics_unittest.cc#L104 | 16:43 |
lambday | added a unit test and the code... although.. the bool param is yet to be added | 16:44 |
lambday | for dense matrices | 16:44 |
lambday | but, this code gives segmentation fault for large dimensions | 16:44 |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 16:44 | |
heiko1 | lambday: good! | 16:46 |
heiko1 | lambday: please dont do new SGMatrix, this produces a memory leak | 16:46 |
heiko1 | just return SGMatrix (S.data ....) | 16:46 |
lambday | heiko1: when I tried to do otherwise, it gave a weird error | 16:46 |
heiko1 | that does it | 16:46 |
lambday | ohkay | 16:46 |
heiko1 | and another thing | 16:46 |
heiko1 | could you generate column vectors? | 16:47 |
heiko1 | as this is the shogun way to store them | 16:47 |
heiko1 | so the result matrix is DxN | 16:47 |
heiko1 | and set N=1 as default | 16:47 |
heiko1 | lambday: you can send the PR, I will check the rest in there (easier to comment) | 16:48 |
lambday | heiko1: okay... but then matrix_mean ias to be given "col_wise = false" right? | 16:48 |
-!- sonne|work [~sonnenbu@sams-office-nat.tomtomgroup.com] has quit [Quit: Leaving.] | 16:48 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 16:50 | |
lambday | heiko1: okay.. just fixing the "new" in SGMatrix and sending the PR | 16:50 |
lambday | heiko1: when I don't use "new", it produces an error.. :( | 16:56 |
lambday | Invalid free() / delete / delete[] / realloc() | 16:58 |
lambday | sorry my mistake | 17:00 |
-!- foulwall [~foulwall@2001:da8:215:c252:6d16:f02f:3a3a:1c5f] has joined #shogun | 17:02 | |
-!- hoijui [~hoijui@dslb-088-074-124-102.pools.arcor-ip.net] has quit [Ping timeout: 264 seconds] | 17:04 | |
-!- lambday [0e8b6149@gateway/web/freenode/ip.14.139.97.73] has quit [Ping timeout: 245 seconds] | 17:09 | |
-!- medeeiip__ [~medeeiip@103.2.133.140] has quit [Read error: Connection reset by peer] | 17:09 | |
-!- medeeiip__ [~medeeiip@103.2.133.140] has joined #shogun | 17:09 | |
-!- lambday [0e8b6149@gateway/web/freenode/ip.14.139.97.73] has joined #shogun | 17:13 | |
-!- abinashpanda [012612ec@gateway/web/freenode/ip.1.38.18.236] has quit [Ping timeout: 245 seconds] | 17:16 | |
-!- mgahlot [~quassel@120.56.198.189] has joined #shogun | 17:17 | |
-!- lambday [0e8b6149@gateway/web/freenode/ip.14.139.97.73] has quit [Ping timeout: 245 seconds] | 17:17 | |
-!- foulwall [~foulwall@2001:da8:215:c252:6d16:f02f:3a3a:1c5f] has quit [Ping timeout: 245 seconds] | 17:36 | |
-!- madhura [~madhura@59.95.233.194] has joined #shogun | 17:40 | |
-!- madhura [~madhura@59.95.233.194] has quit [Ping timeout: 245 seconds] | 17:49 | |
-!- hoijui [~hoijui@wlan-141-23-88-50.tubit.tu-berlin.de] has joined #shogun | 17:50 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Ping timeout: 252 seconds] | 17:57 | |
-!- madhura [~madhura@59.95.232.11] has joined #shogun | 18:02 | |
-!- madhura [~madhura@59.95.232.11] has quit [Ping timeout: 245 seconds] | 18:07 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:13 | |
-!- foulwall_ [~foulwall@2001:da8:215:c252:6d3f:f66:3821:116] has joined #shogun | 18:15 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 18:18 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 18:20 | |
-!- madhura [~madhura@59.95.233.129] has joined #shogun | 18:20 | |
-!- madhura [~madhura@59.95.233.129] has quit [Quit: Leaving] | 18:28 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has joined #shogun | 18:37 | |
-!- foulwall_ [~foulwall@2001:da8:215:c252:6d3f:f66:3821:116] has quit [Ping timeout: 264 seconds] | 18:37 | |
-!- medeeiip [~medeeiip@103.2.133.140] has joined #shogun | 18:37 | |
foulwall | just modified the svr demo. the dots will no longer be saved on the server, so every time enter the page the dots are prepared dots, not the dots inputed last time. And all django views are merged into one function, with only one html template, no data model needed. the demo will be very easy to merge into shogun-web. | 18:38 |
foulwall | sonney2k: | 18:39 |
foulwall | I'll have a look at shogun-web. | 18:40 |
-!- medeeiip__ [~medeeiip@103.2.133.140] has quit [Ping timeout: 252 seconds] | 18:40 | |
foulwall | not so late... | 18:40 |
-!- foulwall_ [~foulwall@li379-21.members.linode.com] has joined #shogun | 18:45 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has quit [Ping timeout: 245 seconds] | 18:46 | |
-!- foulwall_ [~foulwall@li379-21.members.linode.com] has quit [Remote host closed the connection] | 18:50 | |
-!- hoijui [~hoijui@wlan-141-23-88-50.tubit.tu-berlin.de] has quit [Ping timeout: 245 seconds] | 18:53 | |
-!- mgahlot|away_ [~quassel@120.56.198.189] has joined #shogun | 18:55 | |
-!- mgahlot|away__ [~quassel@120.56.198.189] has joined #shogun | 18:56 | |
-!- mgahlot|away__ [~quassel@120.56.198.189] has quit [Read error: Connection reset by peer] | 18:57 | |
-!- mgahlot|away__ [~quassel@120.56.198.189] has joined #shogun | 18:58 | |
-!- mgahlot [~quassel@120.56.198.189] has quit [Ping timeout: 245 seconds] | 18:58 | |
-!- mgahlot|away [~quassel@120.56.198.189] has quit [Ping timeout: 260 seconds] | 18:59 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has joined #shogun | 19:03 | |
foulwall | there's some differences between the site in shogun-web repo and the running one.... Where's the weblog in the navigator bar? | 19:05 |
foulwall | @sonney2k maybe we can add a new url router to host the django demos. | 19:06 |
-!- iglesiasg [~nando@s83-179-44-135.cust.tele2.se] has joined #shogun | 19:06 | |
foulwall | hi iglesiasg | 19:07 |
iglesiasg | hey foulwall | 19:07 |
iglesiasg | how are you doing? | 19:07 |
iglesiasg | I read in logs you were a bit sick | 19:07 |
foulwall | fine! | 19:07 |
iglesiasg | feeling better? | 19:07 |
foulwall | yeah. thanks for caring. | 19:07 |
iglesiasg | no problem :) | 19:07 |
foulwall | I've checkout the shogun-web repo | 19:08 |
iglesiasg | good to hear it is better | 19:08 |
iglesiasg | foulwall: aham ok | 19:08 |
iglesiasg | foulwall: you need admin pass? | 19:08 |
foulwall | aha? | 19:08 |
iglesiasg | sorry? | 19:09 |
foulwall | which admin pass? | 19:09 |
iglesiasg | for the admin interface, but if it was not that, never mind :) | 19:09 |
foulwall | the site in repo or the site running? | 19:09 |
iglesiasg | for the site in the repo I meant | 19:10 |
iglesiasg | the one you run locally for testing | 19:10 |
foulwall | demo needs no data model. so there's no need for the admin panel | 19:11 |
iglesiasg | oh ok, I understand | 19:12 |
foulwall | what i need is a url router to redirect to my demo. | 19:13 |
-!- mgahlot|away__ [~quassel@120.56.198.189] has quit [Remote host closed the connection] | 19:13 | |
iglesiasg | mmm ok, I do not really understand what that means exactly honestly | 19:14 |
-!- mgahlot [~quassel@120.56.198.189] has joined #shogun | 19:14 | |
foulwall | mmm, just I need add rule to make a particular url, like "shogun-toolbox.org/demos/xxxx" connect with my demo | 19:18 |
foulwall | iglesiasg: sorry for my poor english. | 19:18 |
iglesiasg | foulwall: It is not a problem, don't worry for that | 19:18 |
iglesiasg | foulwall: ok, I think I understand what this rule is now | 19:19 |
foulwall | that is called a url router | 19:19 |
foulwall | and each request for that url is passed to their corresponding python function | 19:20 |
foulwall | late now, see you tomorrow, iglesiasg | 19:22 |
iglesiasg | foulwall: ok, see you | 19:23 |
-!- mgahlot [~quassel@120.56.198.189] has quit [Read error: Connection reset by peer] | 19:32 | |
-!- mgahlot|away_ [~quassel@120.56.198.189] has quit [Read error: Connection reset by peer] | 19:32 | |
-!- akshaydixi [~akshay@115.248.50.23] has joined #shogun | 19:42 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:48 | |
-!- foulwall [~foulwall@li379-21.members.linode.com] has quit [Remote host closed the connection] | 19:48 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:49 | |
-!- mgahlot [~quassel@120.56.198.189] has joined #shogun | 19:52 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 19:52 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:53 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Read error: Connection reset by peer] | 19:57 | |
-!- mgahlot is now known as mgahlot|away | 19:58 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 19:58 | |
-!- lisitsyn [~blackburn@109-226-120-160.clients.tlt.100megabit.ru] has joined #shogun | 20:02 | |
-!- mode/#shogun [+o lisitsyn] by ChanServ | 20:02 | |
-!- mgahlot [~quassel@120.56.198.189] has joined #shogun | 20:03 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:03 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:04 | |
-!- lisitsyn [~blackburn@109-226-120-160.clients.tlt.100megabit.ru] has quit [Client Quit] | 20:04 | |
-!- lisitsyn [~blackburn@109-226-120-160.clients.tlt.100megabit.ru] has joined #shogun | 20:05 | |
-!- mode/#shogun [+o lisitsyn] by ChanServ | 20:05 | |
-!- iglesiasg [~nando@s83-179-44-135.cust.tele2.se] has quit [Quit: leaving] | 20:05 | |
-!- mgahlot|away [~quassel@120.56.198.189] has quit [Ping timeout: 245 seconds] | 20:06 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:07 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:08 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:11 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:12 | |
-!- heiko1 [~heiko@nat-166-51.internal.eduroam.ucl.ac.uk] has left #shogun [] | 20:14 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:15 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:16 | |
-!- medeeiip [~medeeiip@103.2.133.140] has quit [Quit: Leaving] | 20:18 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:19 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:20 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:23 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:24 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:27 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has joined #shogun | 20:28 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:28 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:31 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:32 | |
-!- medeeiip [~medeeiip@103.2.133.140] has joined #shogun | 20:32 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:35 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:36 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:39 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:40 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:43 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:44 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:47 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:48 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:51 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:52 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:55 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 20:57 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 20:59 | |
-!- medeeiip__ [~medeeiip@103.2.133.140] has joined #shogun | 21:00 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:01 | |
-!- medeeiip [~medeeiip@103.2.133.140] has quit [Ping timeout: 245 seconds] | 21:03 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:04 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:05 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:08 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:09 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:12 | |
-!- akshaydixi [~akshay@115.248.50.23] has quit [Ping timeout: 245 seconds] | 21:13 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:13 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:16 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:17 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:20 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:21 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:24 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:25 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:28 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:29 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:32 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:33 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:36 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:37 | |
-!- mgahlot [~quassel@120.56.198.189] has quit [Read error: Connection reset by peer] | 21:39 | |
-!- akshaydixi [~akshay@112.79.42.186] has joined #shogun | 21:41 | |
-!- mgahlot [~quassel@120.56.198.189] has joined #shogun | 21:41 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:41 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:42 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:47 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:48 | |
-!- jptech93 [~quassel@115.248.130.148] has quit [Remote host closed the connection] | 21:53 | |
-!- jptech93 [~quassel@115.248.130.148] has joined #shogun | 21:55 | |
-!- sorcerer [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #shogun | 22:01 | |
-!- sumit [73f91219@gateway/web/freenode/ip.115.249.18.25] has quit [Ping timeout: 245 seconds] | 22:07 | |
-!- zxtx [~zv@cpe-75-83-151-252.socal.res.rr.com] has quit [Ping timeout: 264 seconds] | 22:14 | |
-!- mgahlot|away [~quassel@120.56.205.144] has joined #shogun | 22:22 | |
-!- mgahlot [~quassel@120.56.198.189] has quit [Ping timeout: 276 seconds] | 22:26 | |
-!- akshaydixi [~akshay@112.79.42.186] has quit [Ping timeout: 252 seconds] | 22:30 | |
-!- sorcerer [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Ping timeout: 245 seconds] | 22:38 | |
-!- akshaydixi [~akshay@223.233.3.87] has joined #shogun | 22:43 | |
-!- mgahlot|away [~quassel@120.56.205.144] has quit [Ping timeout: 276 seconds] | 23:06 | |
-!- akshaydixi [~akshay@223.233.3.87] has quit [Ping timeout: 252 seconds] | 23:16 | |
@lisitsyn | sonney2k: nice | 23:17 |
@sonney2k | lisitsyn, that should be enough spam | 23:17 |
@lisitsyn | sonney2k: who are connectionists? | 23:18 |
@sonney2k | lisitsyn, now we should do the workshop announcement | 23:18 |
@sonney2k | lisitsyn, I would rather want to do a broad announcement | 23:18 |
@sonney2k | like "shogun machine learning workshop" | 23:18 |
@sonney2k | and then say that it will feature general (applied) machine learning talks | 23:19 |
@sonney2k | with particular focus on the shogun machine learning toolbox | 23:19 |
@lisitsyn | sonney2k: makes sense | 23:19 |
@sonney2k | so we could have some people just generally interested in ML | 23:19 |
@lisitsyn | sonney2k: I should dig into books to learn something on machine learning then! | 23:19 |
@lisitsyn | :D | 23:19 |
@sonney2k | lisitsyn, connectionists are the root of it all | 23:20 |
@sonney2k | the neural net guys / nips roots etc | 23:20 |
@lisitsyn | sonney2k: I am ashamed I didn't know about that list | 23:20 |
@lisitsyn | sonney2k: do they have a lot of discussions there? | 23:20 |
@sonney2k | lisitsyn, I like ml-news better these days | 23:21 |
@lisitsyn | embed(...,(method=LaplacianEigenmaps,eigen_method=Randomized)) | 23:22 |
@lisitsyn | sonney2k: ^ it works :D | 23:22 |
@sonney2k | :) | 23:24 |
@sonney2k | I just realize I have to write sth to gunnar | 23:25 |
@sonney2k | *argh* | 23:25 |
@lisitsyn | sonney2k: did he agree? | 23:25 |
-!- mgahlot [~quassel@59.177.128.177] has joined #shogun | 23:28 | |
-!- akshaydixi [~akshay@223.189.62.148] has joined #shogun | 23:30 | |
@sonney2k | lisitsyn, chris said so | 23:31 |
@sonney2k | but that is about sth else | 23:31 |
@lisitsyn | sonney2k: gunnar looks like a bigger big shot than obama | 23:31 |
@sonney2k | this year gsoc is crazy - I mean it really ends late september | 23:32 |
@sonney2k | and already did start now | 23:32 |
@lisitsyn | yees | 23:32 |
@sonney2k | it is like 6 months | 23:32 |
@lisitsyn | I think we will all die | 23:32 |
@sonney2k | lisitsyn, now or never ;) | 23:32 |
@sonney2k | but ultimately I agree | 23:33 |
@lisitsyn | sonney2k: our students will die too | 23:33 |
@lisitsyn | endless summer | 23:34 |
@lisitsyn | sonney2k: more connectionists? | 23:37 |
@sonney2k | maybe it gives them the freedom to manage both studies & gsoc | 23:39 |
@sonney2k | this stretching reduces and increases pressure at the same time | 23:40 |
@sonney2k | lisitsyn, yes connectionists did not allow me to post from this address | 23:51 |
-!- akshaydixi [~akshay@223.189.62.148] has quit [Read error: Connection reset by peer] | 23:51 | |
--- Log closed Sat Apr 13 00:00:27 2013 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!