--- Log opened Mon Jun 04 00:00:35 2018 | ||
-!- witness_ [uid10044@gateway/web/irccloud.com/x-tudoqcrrhfhjtitk] has quit [Quit: Connection closed for inactivity] | 00:15 | |
-!- witness_ [uid10044@gateway/web/irccloud.com/x-gkqmyuxkwjzcztje] has joined #shogun | 02:53 | |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4326 opened by vinx13 | 05:25 |
---|---|---|
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 08:47 | |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 09:00 | |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 09:02 | |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 09:23 | |
@sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4308 closed by vigsterkr | 11:38 |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4325 merged by vigsterkr | 11:38 |
@sukey | [https://github.com/shogun-toolbox/shogun] New commit https://github.com/shogun-toolbox/shogun/commit/cd08c4ee16dc02e9d64e3ba7be7fcb734de0f0de by vigsterkr | 11:38 |
-shogun-buildbot:#shogun- Build deb4 - python3 #444 is complete: Failure [failed python (failure)] - http://buildbot.shogun-toolbox.org:8080/#builders/28/builds/444 | 11:49 | |
-!- HeikoS [~heiko@host86-128-122-106.range86-128.btcentralplus.com] has joined #shogun | 11:55 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 11:55 | |
--- Log closed Mon Jun 04 12:15:55 2018 | ||
--- Log opened Mon Jun 04 12:16:05 2018 | ||
-!- shogun-t1olbox [~shogun@7nn.de] has joined #shogun | 12:16 | |
-!- Irssi: #shogun: Total of 17 nicks [3 ops, 0 halfops, 0 voices, 14 normal] | 12:16 | |
-!- Irssi: Join to #shogun was synced in 9 secs | 12:16 | |
-!- Netsplit *.net <-> *.split quits: shogun-toolbox | 12:21 | |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4305 synchronized by shubham808 | 12:39 |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 12:42 | |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4327 opened by shubham808 | 12:52 |
-!- witness_ [uid10044@gateway/web/irccloud.com/x-gkqmyuxkwjzcztje] has quit [Quit: Connection closed for inactivity] | 12:54 | |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 13:47 | |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 13:47 | |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Client Quit] | 13:49 | |
-!- Farouk [9cc52ce4@gateway/web/freenode/ip.156.197.44.228] has joined #shogun | 14:27 | |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4206 synchronized by FaroukY | 14:35 |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4206 synchronized by FaroukY | 14:38 |
Farouk | wiking, you here? | 14:41 |
wuwei | wiking: hi | 15:10 |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4326 synchronized by vinx13 | 15:14 |
-!- HeikoS [~heiko@eduroam-int-pat-8-0.ucl.ac.uk] has joined #shogun | 15:44 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 15:44 | |
-!- HeikoS [~heiko@eduroam-int-pat-8-0.ucl.ac.uk] has quit [Ping timeout: 260 seconds] | 15:49 | |
lisitsyn | what's going on :P | 15:51 |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 15:57 | |
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun | 16:03 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 16:03 | |
Farouk | HeikoS: Hi | 16:19 |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 16:20 | |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has joined #shogun | 16:22 | |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4317 synchronized by vinx13 | 16:23 |
@HeikoS | Farouk: hihi | 16:27 |
@HeikoS | how are things? | 16:27 |
@HeikoS | shubham808: hi | 16:27 |
shubham808 | HeikoS: hey | 16:27 |
Farouk | good good. working on porting SGDMinimizer to use stan now. However, I have a question on the cookbook. After I updated my shogun-data, I noticed binary_perceptron.dat was changed in shogun data | 16:28 |
Farouk | even though I never touched it | 16:28 |
@HeikoS | Farouk: there are two ways to change it | 16:29 |
@HeikoS | updating the data submodule | 16:29 |
@HeikoS | and changing the file | 16:29 |
@HeikoS | Farouk: but I think you can just ignore this or? | 16:29 |
Farouk | Not sure, because it caused a conflict in the branch. | 16:30 |
Farouk | Its already merged in shogun-data | 16:30 |
Farouk | the cross_validation.dat | 16:30 |
@HeikoS | I am not sure what you are trying to do or what you are being blocked | 16:31 |
@HeikoS | but you can always git reset the data submodule | 16:31 |
@HeikoS | and then do your changes again | 16:31 |
Farouk | Alright I'll try that | 16:32 |
shubham808 | HeikoS: hey | 16:33 |
@HeikoS | Farouk: let me know if you have problems | 16:33 |
@HeikoS | should not block you in any way | 16:33 |
@HeikoS | if it does, ask again :) | 16:33 |
@HeikoS | shubham808: whats up? | 16:33 |
shubham808 | about the garbage collection pr | 16:33 |
shubham808 | valgrind ctest -R generated* --without-pymalloc --leak-check=full | 16:34 |
shubham808 | does not detect anything | 16:34 |
@HeikoS | shubham808: and how does that change from before? | 16:35 |
@HeikoS | before the patch I mean | 16:35 |
shubham808 | i have to check with the pymalloc flag | 16:36 |
shubham808 | (for before the patch) | 16:36 |
@HeikoS | kk | 16:38 |
shubham808 | another thing | 16:38 |
shubham808 | for feature block logistic regression cookbook | 16:39 |
shubham808 | i am stuck at a place | 16:39 |
@HeikoS | yep | 16:40 |
@HeikoS | have something I can look at? | 16:40 |
shubham808 | https://github.com/shubham808/shogun/blob/fba2d9ec06e545f42fb1b3f212f6956b1f7e0585/examples/meta/src/binary/feature_block_logistic_regression.sg#L21 | 16:40 |
@HeikoS | Farouk: I now see your conflict | 16:41 |
@HeikoS | so what you need to do | 16:41 |
@HeikoS | 1.) locally reset your data repo to master | 16:41 |
@HeikoS | 2.) generate the file from your example and copy it into the data repo | 16:41 |
Farouk | Alright I see | 16:42 |
@HeikoS | (or actually, since wiking has already merged the file, no need to do that if you have resetted it to master) | 16:42 |
@HeikoS | 3.) commit the data version in your 4206 PR | 16:42 |
Farouk | I did, when travis tests, does it install the .dat files from shogun-data? | 16:42 |
@HeikoS | then the conflict will be gone | 16:42 |
@HeikoS | yes it does | 16:42 |
@HeikoS | it pulls the commit with the hash that is in the submodule shogun/data | 16:42 |
Farouk | Ohhh I see | 16:43 |
@HeikoS | Farouk: you can see that in the travis logs | 16:43 |
@HeikoS | https://travis-ci.org/shogun-toolbox/shogun/jobs/386990066#L446 | 16:43 |
@HeikoS | eg here | 16:43 |
@HeikoS | it tells you exactly which commit it checks out | 16:43 |
@HeikoS | and that is the same as in the submodule that you update in the main repo | 16:43 |
@HeikoS | Farouk: let's aim to merge this now | 16:44 |
Farouk | Alright I'll give it a go now | 16:44 |
@HeikoS | Farouk: as in the next few minutes | 16:44 |
Farouk | I am pausing other work | 16:44 |
@HeikoS | as this has been open to long :) | 16:44 |
Farouk | Yep, I agree :) | 16:45 |
@HeikoS | so give it a try and let me know what happens | 16:45 |
Farouk | Sure, trying now | 16:45 |
-!- shubham808 [0e8bf0fb@gateway/web/cgi-irc/kiwiirc.com/ip.14.139.240.251] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] | 16:46 | |
shogitter | (shubham808) Heiko hey I switched to shogitter | 16:46 |
shogitter | (shubham808) Something went wrong on my pc | 16:47 |
shogitter | (shubham808) U see the line of code above ? | 16:48 |
@HeikoS | lol ok | 16:48 |
@HeikoS | checking the code | 16:49 |
@HeikoS | yes the factory | 16:49 |
@HeikoS | whats the prob? | 16:49 |
shogitter | (shubham808) So the feature_relation is expects a indexblockrelation pointer | 16:49 |
shogitter | (shubham808) But what we are passing is some<indexblockgroup> | 16:50 |
shogitter | (shubham808) U see the problem? | 16:51 |
@HeikoS | not yet | 16:51 |
@HeikoS | ah | 16:51 |
@HeikoS | base class vs specialised class | 16:51 |
shogitter | (shubham808) Yep | 16:51 |
@HeikoS | ah ok | 16:51 |
shogitter | (shubham808) I tried adding a new base class | 16:51 |
@HeikoS | sure that needs fixing for meta examples | 16:51 |
@HeikoS | everything should be done via base class | 16:51 |
shogitter | (shubham808) Yes yes | 16:51 |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4206 synchronized by FaroukY | 16:51 |
shogitter | (shubham808) But then I won't be able to use block_add(block1) anymore | 16:52 |
@HeikoS | so you need a factory that instantiates the group | 16:52 |
@HeikoS | uh | 16:52 |
@HeikoS | because that is specialised method? | 16:52 |
shogitter | (shubham808) Yes | 16:52 |
@HeikoS | i see | 16:52 |
@HeikoS | let me think a moment | 16:52 |
@HeikoS | interesting problem | 16:53 |
shogitter | (shubham808) So then I tried to do a put_add thingy we did for combined kernel | 16:53 |
@HeikoS | yes | 16:53 |
@HeikoS | that would be my first relfex as well | 16:53 |
shogitter | (shubham808) But that would mean making block group a base too | 16:53 |
@HeikoS | but we have add only for objects | 16:53 |
shogitter | (shubham808) Since only base can be "added" to base | 16:53 |
@HeikoS | let me stare at the example for a bit | 16:53 |
shogitter | (shubham808) Okay ? | 16:54 |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4317 synchronized by vinx13 | 16:55 |
shogitter | (shubham808) Hey I will get set up for meeting let's take this up there | 16:55 |
@HeikoS | ok | 16:55 |
@HeikoS | see you in 5 | 16:55 |
@HeikoS | I attached a hangout call to the invite | 16:55 |
@HeikoS | let's meet in there | 16:55 |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4206 synchronized by FaroukY | 16:55 |
@HeikoS | lisitsyn: you there? | 16:56 |
shogitter | (shubham808) In 10 ? my pc is still weird | 16:56 |
@HeikoS | ok :D | 16:56 |
@HeikoS | Farouk: still conflicting | 16:57 |
@HeikoS | you need to make sure your data submodule is resetted to the shogun-data/master head | 16:57 |
@HeikoS | you branched at some point | 16:57 |
Farouk | HeikoS: I think the problem is that there is another pr that is changing averaged_perceptron.dat and perceptron.dat | 16:57 |
Farouk | I did | 16:57 |
@sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4328 vigsterkr added label: "SWIG" | 16:57 |
@sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4328 | 16:57 |
Farouk | its resetted to shogun-data/head | 16:57 |
@sukey | [https://github.com/shogun-toolbox/shogun] Issue https://github.com/shogun-toolbox/shogun/issues/4328 opened by vigsterkr | 16:57 |
@HeikoS | Farouk: no thats not the problem | 16:57 |
@HeikoS | it is a simple git merge conflict | 16:58 |
@HeikoS | which means you are in a different branch than the develop branch | 16:58 |
@HeikoS | you will need to fully regenerate your commit with the .dat file | 16:58 |
@HeikoS | or merge the conflict locally | 16:58 |
rcurtin | HeikoS: ping re: MLOSS workshop :) | 17:00 |
rcurtin | deadline is the 11th, just wanted to see what you were thinking | 17:00 |
@wiking | :) | 17:01 |
@wiking | rcurtin, morning :) | 17:01 |
wuwei | wiking: hi | 17:01 |
rcurtin | morning to you also... although it is almost lunchtime here. I am not the earliest riser in the world... | 17:01 |
@wiking | wuwei, yoyoyo | 17:01 |
@wiking | rcurtin, :> | 17:01 |
wuwei | wiking: have u seen the broken tests of my cookbook pr? because transformer branch is out of date | 17:02 |
wuwei | wiking: shall we have a new branch for transformer in shogun-data? | 17:03 |
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 256 seconds] | 17:03 | |
@wiking | mmmm we ccould | 17:03 |
@wiking | but first i wanna rebase that branch :) | 17:03 |
@wiking | lemme merge the inverse transform bracnh | 17:04 |
@wiking | and rebase the feature branch | 17:04 |
@wiking | ok? | 17:04 |
wuwei | yeah sure | 17:04 |
@sukey | [https://github.com/shogun-toolbox/shogun] vigsterkr pushed 4 commits: | 17:04 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/41659b8274cbbf735e75c6170205f00e544f4cba | 17:04 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/b45c31eb0407c7a60760dfd69e08389fda16ad9a | 17:04 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/84c1cde758d1a8752fbcc14694464c58f6177eb2 | 17:04 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/45162b6952051839f8a908983f6100cea0805821 | 17:04 |
@sukey | [https://github.com/shogun-toolbox/shogun] Pull Request https://github.com/shogun-toolbox/shogun/pull/4326 merged by vigsterkr | 17:04 |
@wiking | hehehe rebase problem | 17:05 |
@wiking | lemme resolve it | 17:05 |
@wiking | :) | 17:05 |
wuwei | well.. have you seen the one failing test in inverse_transform | 17:06 |
@wiking | noup | 17:06 |
wuwei | a bit weird, not sure about that | 17:07 |
@sukey | [https://github.com/shogun-toolbox/shogun] vigsterkr pushed 67 commits: | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/5766fd9438ea89c9dea641cdd3970badae74b2e6 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/bf0e5c82bf1bfb9d5e1424ab70bd25d709100d83 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/c13a5f3b06ac3f2b2d19342ae6d61dd2e42d4bd5 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/46eab927c4f425c4bf4687fc2626c51fd93d0d06 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/727384a6dc8ee66177a3ffae2276086b7faf5609 | 17:10 |
@wiking | wuwei, ^ rebased the feature branch | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/0deb4c6d280eec57cbb0ce3389aae381d850d9a3 | 17:10 |
@wiking | :) | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/23c34e9f210df1f24def971d61db156c4990386f | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/ca8397d9199648fcff2ce0e07c8b30007a7d2d6d | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/eb3c7af4579f523cc99922c1841c5faef455752a | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/a1084d46b627bacde66e513e3292890128591d92 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/88a110c89ba5982178f5b4772766d2749c4dbe37 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/b243db9a871ca0673cd0dc83fc1377d81c65c4f5 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/bf4b21f36c241cb52ce002651740ede6a7514fc4 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/b4ed9e01110b740256ee3193be9e27ecf95de7a9 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/080e2d776b0e305f9c2f5c6aca6e3c0e9bb25bad | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/e30eba40394e8621d3e71903591fdf28a7a52305 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/f5638480f2147734eee67321f0ace8b7b524547e | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/a8c232d7b07a2d6c8f435e71a6dbe5b188cf16bd | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/bd6fbb6c9e1b039032d8d964edc835d470a95f3b | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/b3ac8ec37692482cb809985966c9252cc1c84299 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/8a3e4d4346a344cf33798dfd5ca583576d61021b | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/11e72d664201ec28000d37b0b77215da80b464f3 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/4b6f6bbb61aba36de9fc4c20370f6953a1aa7aab | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/b9f907ea6daa67bedb708dbabe03282fb49a7075 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/752aab19a7bb40aa80c62985c61418a345e21538 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/788b1ccc8701ed9ac13066f9b6e04cb0b6bd903c | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/d26eca7073371906c7cf320ce2712075c35246f0 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/9d875ede5ce6399cdaf702143428939f84848b91 | 17:10 |
@sukey | https://github.com/shogun-toolbox/shogun/commit/ba362a027b79247133d4acf47331deff10e5202d | 17:10 |
-!- sukey [~nodebot@ks312251.kimsufi.com] has quit [Remote host closed the connection] | 17:10 | |
wuwei | great, thanks! | 17:10 |
@wiking | so you'l have to rebase https://github.com/shogun-toolbox/shogun/pull/4317 | 17:11 |
@wiking | and the other stuff :))) | 17:11 |
Farouk | HeikoS: Alright conflict fixed | 17:16 |
-!- travis-ci [~travis-ci@ec2-54-92-224-122.compute-1.amazonaws.com] has joined #shogun | 17:18 | |
travis-ci | it's Wuwei Lin's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/387836355 | 17:18 |
-!- travis-ci [~travis-ci@ec2-54-92-224-122.compute-1.amazonaws.com] has left #shogun [] | 17:18 | |
-!- travis-ci [~travis-ci@ec2-54-242-138-74.compute-1.amazonaws.com] has joined #shogun | 17:19 | |
travis-ci | it's Wuwei Lin's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/387836355 | 17:19 |
-!- travis-ci [~travis-ci@ec2-54-242-138-74.compute-1.amazonaws.com] has left #shogun [] | 17:19 | |
-!- HeikoS [~heiko@eduroam-int-pat-8-0.ucl.ac.uk] has joined #shogun | 17:19 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 17:19 | |
@wiking | wuwei, ping | 17:34 |
wuwei | wiking: hi | 17:35 |
@wiking | wuwei, can u do a quick test with a swig interface (say python) that if you throw from a pipeline an std::invalid_argument what happens | 17:35 |
@wiking | (use python shell to do the test) | 17:35 |
@wiking | i would think that in that you will exit from python shell | 17:35 |
@wiking | but maybe i'm wrong | 17:35 |
@wiking | in case of ShogunException that does not happen | 17:36 |
wuwei | sure | 17:36 |
@wiking | i hope this will kick in | 17:40 |
@wiking | https://github.com/swig/swig/blob/master/Lib/typemaps/exception.swg | 17:40 |
@wiking | https://github.com/swig/swig/blob/master/Lib/typemaps/exception.swg#L60 | 17:40 |
@wiking | but lets just check | 17:42 |
-!- Farouk [9cc52ce4@gateway/web/freenode/ip.156.197.44.228] has quit [Ping timeout: 260 seconds] | 17:42 | |
-!- Farouk [9cc52ce4@gateway/web/freenode/ip.156.197.44.228] has joined #shogun | 17:52 | |
wuwei | wiking: python exited after exception | 17:59 |
wuwei | terminate called after throwing an instance of 'std::invalid_argument' | 17:59 |
wuwei | what(): error | 17:59 |
wuwei | [1] 14037 abort (core dumped) python | 17:59 |
@HeikoS | wiking: jojo | 18:09 |
@HeikoS | wiking: there is some discussion necessary regarding black box stuff | 18:10 |
@HeikoS | regarding what makes the state of the model | 18:10 |
@HeikoS | and whether it might be good to have some kind of internal API to update model state | 18:10 |
@HeikoS | since otherwise it is hard to expose/catch | 18:10 |
@HeikoS | things like | 18:11 |
@HeikoS | number of iterations | 18:11 |
@HeikoS | are usually just temporary variables | 18:11 |
@HeikoS | so exposing them (say to progress bar) happens manually | 18:11 |
@HeikoS | for iterations it might not be a big issue | 18:11 |
@HeikoS | but there might be cases where temporary variables are an issue | 18:11 |
@HeikoS | next thing is that we somehow want to ensure that stopped models are seirlilizable with the temporary state they are in | 18:12 |
@HeikoS | however, that is basically impossible to test if the train method just directly updates state variables (or not) | 18:12 |
@HeikoS | if there was some kind of API like | 18:12 |
@HeikoS | to update the state in an iteration | 18:13 |
@HeikoS | then one could hitchhike this and connect things for testing and exposing these numbers | 18:13 |
@HeikoS | maybe we can discuss this tomorrow at stammtisch | 18:13 |
@HeikoS | Farouk: how is that data PR giong? :D | 18:13 |
-!- HeikoS [~heiko@eduroam-int-pat-8-0.ucl.ac.uk] has quit [Ping timeout: 240 seconds] | 18:17 | |
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has joined #shogun | 18:31 | |
-!- mode/#shogun [+o HeikoS] by ChanServ | 18:31 | |
lisitsyn | HeikoS: hey | 18:56 |
@HeikoS | lisitsyn: hi! | 18:56 |
@HeikoS | lisitsyn: I have a few qs for you | 18:56 |
@HeikoS | have some time? | 18:56 |
lisitsyn | yeah | 18:56 |
lisitsyn | I am back | 18:57 |
@HeikoS | cool | 18:58 |
@HeikoS | sooo | 19:02 |
@HeikoS | lisitsyn: first thing: https://github.com/shubham808/shogun/blob/fba2d9ec06e545f42fb1b3f212f6956b1f7e0585/examples/meta/src/binary/feature_block_logistic_regression.sg#L21 | 19:02 |
@HeikoS | lisitsyn: your index block API is quite hard to represent with factories and base classes | 19:02 |
@HeikoS | as in order to build them, one needs to call methods on specialised class | 19:03 |
@HeikoS | we could have another "add" extension, but I wonder whether an OO interface is the right thing here | 19:03 |
@HeikoS | maybe rather index based? | 19:03 |
Farouk | HeikoS: It's done now, fixed the conflict | 19:08 |
Farouk | Waiting for travis | 19:09 |
-!- travis-ci [~travis-ci@ec2-54-242-138-74.compute-1.amazonaws.com] has joined #shogun | 19:10 | |
travis-ci | it's Wuwei Lin's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: https://travis-ci.org/shogun-toolbox/shogun/builds/387839200 | 19:10 |
-!- travis-ci [~travis-ci@ec2-54-242-138-74.compute-1.amazonaws.com] has left #shogun [] | 19:10 | |
@HeikoS | Farouk: cool nice one :) | 19:10 |
Farouk | Turns out like you said, I had committed the head before, so I just undid that commit, then reset data head, then commit again, and it was good | 19:11 |
@HeikoS | nice | 19:11 |
@HeikoS | something learnt about git :) | 19:11 |
Farouk | Yeah :) | 19:11 |
rcurtin | HeikoS: thanks for the email update :) | 19:12 |
@HeikoS | rcurtin: welcome :) | 19:13 |
@HeikoS | rcurtin: lisitsyn suggested another speaker from yandex | 19:13 |
@HeikoS | but I will wait for the other ones to get back first | 19:13 |
rcurtin | sure, no problem, and if you'd like me to reach out to anyone just let me know | 19:14 |
@HeikoS | cool | 19:16 |
@HeikoS | rcurtin: how is gsoc going? | 19:17 |
rcurtin | it's going well so far, this year we have six students | 19:17 |
rcurtin | so far less insane than last year's 10 students | 19:17 |
rcurtin | by the way, I also know Misha Bilenko at Yandex, he keeps saying hi to me at conferences. he led the development of catboost, so we could also reach out to him (and maybe also Anna too), your call | 19:18 |
@HeikoS | that is quite inline with what sergey suggested, so its on the "try next" list :) | 19:19 |
rcurtin | John Langford I think would be a good person that people would show up for, but I don't know if he'd accept. he is pretty busy so he may have other things to do | 19:20 |
rcurtin | but I guess we only need a couple speakers, not 20 :) | 19:20 |
@HeikoS | haha yes indeed | 19:21 |
lisitsyn | bilenko is a big shot :) | 19:25 |
lisitsyn | HeikoS: let me take a look | 19:26 |
-!- HeikoS [~heiko@untrust-out.swc.ucl.ac.uk] has quit [Ping timeout: 265 seconds] | 19:26 | |
-!- witness_ [uid10044@gateway/web/irccloud.com/x-ochxnustouiwkdtw] has joined #shogun | 20:45 | |
-!- witness_ [uid10044@gateway/web/irccloud.com/x-ochxnustouiwkdtw] has quit [Quit: Connection closed for inactivity] | 23:25 | |
--- Log closed Tue Jun 05 00:00:36 2018 |
Generated by irclog2html.py 2.10.0 by Marius Gedminas - find it at mg.pov.lt!