[07:14:23] 10Lift-Wing, 10artificial-intelligence, 10articlequality-modeling, 10Machine-Learning-Team (Active Tasks): Upload articlequality model binaries to storage - https://phabricator.wikimedia.org/T307417 (10kevinbazira) 13/13 articlequality models were uploaded successfully to Thanos Swift. [07:16:30] 10Lift-Wing, 10artificial-intelligence, 10articlequality-modeling, 10Machine-Learning-Team (Active Tasks), 10Patch-For-Review: Create articlequality inference services - https://phabricator.wikimedia.org/T307418 (10kevinbazira) Inference services were created for all the 13 articlequality models and they... [07:19:26] 10Lift-Wing, 10artificial-intelligence, 10articlequality-modeling, 10Machine-Learning-Team (Active Tasks): Migrate articlequality models - https://phabricator.wikimedia.org/T307416 (10kevinbazira) [07:23:26] hello folks [07:25:19] 10Lift-Wing, 10artificial-intelligence, 10articlequality-modeling, 10Machine-Learning-Team (Active Tasks): Migrate articlequality models - https://phabricator.wikimedia.org/T307416 (10kevinbazira) The migration of articlequality models has been completed. 13/13 articlequality [[ https://phabricator.wikime... [09:09:24] hi elukey o/ [09:09:50] Whenever you have a minute to jump onto a quick call please let me know [09:10:28] I'd like to get some clarity on what part of the preprocess should be async [09:12:14] kevinbazira: o [09:12:15] o/ [09:12:39] I have a meeting in ~15 min, if you want we can do something quick now [09:13:19] great calling now ... [09:13:19] but basically the idea would be to have HTTP calls to the MW API done via async http client [09:14:35] kevinbazira: are you going to create a meeting invite? [09:14:40] err sorry a meeting link [09:14:45] otherwise I can create one [09:14:49] meet.google.com/igy-gvzz-kxg [09:14:53] ah ok :) [09:46:48] 10Machine-Learning-Team, 10serviceops: Fix calico, cfssl-issuer and knative-serving Helm dependencies - https://phabricator.wikimedia.org/T303279 (10JMeybohm) [10:24:43] elukey: in case you missed it: I did some cleanup of /srv on the sandbox yesterday. Deleted unused Docker layers, and removed some large repos from andy's homedir (I check and they had no changes/branches that weren't also on github). I also dropped some old versions of the accraze/editquality-aixserver image. We're now at 15G free again. [10:58:36] <<- lunch [11:53:06] 10Machine-Learning-Team, 10Patch-For-Review, 10Platform Team Initiatives (API Gateway), 10Platform Team Workboards (Platform Engineering Reliability): Proposal: add a per-service rate limit setting to API Gateway - https://phabricator.wikimedia.org/T295956 (10hnowlan) One major issue with our previous appr... [13:23:05] klausman: ack yes I've read it! thanks [14:14:24] kevinbazira: do you want to meet again to talk about the async stuff? [14:25:13] still digging into it. will keep you posted. [14:28:45] sure [14:39:32] 10Lift-Wing, 10Machine-Learning-Team (Active Tasks): Test async preprocess on kserve - https://phabricator.wikimedia.org/T309623 (10elukey) On stat1004 I created this small script: ` import revscoring from revscoring import Model from revscoring.extractors import api import mwapi import nltk import logging l... [14:46:33] folks I added some info about the MW API calls that a single rev-id causes (for feature extraction) --^ [14:48:54] :+1: [15:03:04] morning! [15:03:13] \o [15:42:01] 10Lift-Wing, 10Machine-Learning-Team (Active Tasks): Test async preprocess on kserve - https://phabricator.wikimedia.org/T309623 (10kevinbazira) As I was thinking through async, I noticed the feature extraction code is being called twice here: https://github.com/wikimedia/machinelearning-liftwing-inference-ser... [16:05:03] * elukey afk! [16:54:26] ditto, see y'all tomorrow [17:06:03] bye Tobias!