[06:18:30] <wikibugs>	 10Scoring-platform-team (Current), 10Research, 10drafttopic-modeling: Add wikidata to articletopic pipeline - https://phabricator.wikimedia.org/T254289 (10Dibyaaaaax) **Run-5**   | Classifier| Fasttext |Parameters | `loss=ova`  `epoch=25`  `dim=50`  `lr=0.1`  `pretrainedVectors=word2vec/wikidata-20200501-lea...
[10:47:37] <wikibugs>	 10Scoring-platform-team (Current), 10Research, 10drafttopic-modeling: Add wikidata to articletopic pipeline - https://phabricator.wikimedia.org/T254289 (10Dibyaaaaax) **Run-6**   | Classifier| Fasttext |Parameters | `loss=ova`  `epoch=25`  `dim=50`  `lr=0.1`  `pretrainedVectors=word2vec/wikidata-20200501-lea...
[13:51:10] <wikibugs>	 10Scoring-platform-team, 10drafttopic-modeling: Clean up History and Society.Society in the topic taxonomy. - https://phabricator.wikimedia.org/T246912 (10Isaac) See https://github.com/wikimedia/wikitax/pull/6 for implementation of these changes
[14:01:04] <wikibugs>	 10Scoring-platform-team (Current), 10Research, 10drafttopic-modeling: Add wikidata to articletopic pipeline - https://phabricator.wikimedia.org/T254289 (10Dibyaaaaax) **Run-7**   | Classifier| GradientBoosting |Parameters | `n_estimators=150`   `max_depth=5`  `max_features="log2"`   `learning_rate=0.1` | Num...
[15:35:00] <dsaez>	 hi all. Do you any pointer to find the list of features used by ORES to compute articles' quality?
[19:41:31] <mutante>	 hey there, ORES people
[19:42:05] <mutante>	 i am wondering what would be a good test to see if ORES works also over https
[19:42:07] <Zppix>	 Aaron doesnt seem to be connected
[19:42:13] <Zppix>	 kevinbazira: ^
[19:42:17] <Zppix>	 Could you help
[19:44:18] <mutante>	 so.. the goal is: have TLS between caching layer (ATS) and the ORES backends. and what i did is: add envoy-proxy on each ores* backend, create a certificate for, configure it to listen on 443 and talk "upstream" (local_port) to 8081
[19:44:27] <mutante>	 that's where uwsgi is listening
[19:48:18] <mutante>	 what i'm testing is ... is talking to existing "http://ores.discovery.wmnet:8081" now equivalent to talking to "https://ores.discovery.wmnet". and it's not yet, so i will keep debugging that 
[22:00:50] <codezee>	 need some help with `revscoring extract` I'm extracting features for about 1M revisions, but the script terminated in between because of some error. Does anyone know how to resume extract from the point it was terminated?