[08:12:01] 10Machine-Learning-Team, 10SRE, 10serviceops, 10Kubernetes, 10Patch-For-Review: Add the possibility to deploy calico on kubernetes master nodes - https://phabricator.wikimedia.org/T285927 (10elukey) `base::expose_puppet_certs` is used in both master and node profiles, with different settings: * on master... [08:12:35] good morning :) [08:43:19] 10Machine-Learning-Team, 10SRE, 10serviceops, 10Kubernetes, 10Patch-For-Review: Add the possibility to deploy calico on kubernetes master nodes - https://phabricator.wikimedia.org/T285927 (10elukey) After a chat with Janis we reviewed the master's code and found https://gerrit.wikimedia.org/r/c/operation... [11:13:46] I am adding docker + kubelet on ml-serve-ctrl1001, seems working, will keep going later on! [11:13:49] * elukey lunch! [14:04:41] * elukey coffee [15:22:21] morning all [15:22:43] good morning :) [15:23:19] I got into the GitHub Copilot beta so now I never need to code ever again [15:36:42] Nice [15:37:10] It's copilot though not pilot [15:37:23] I've heard many stories about it [15:37:53] Ha, too true [16:16:27] Just make sure to not end up with CFIT :) [17:37:59] 10Machine-Learning-Team, 10SRE, 10serviceops, 10Kubernetes, 10Patch-For-Review: Add the possibility to deploy calico on kubernetes master nodes - https://phabricator.wikimedia.org/T285927 (10elukey) Kubelets / calico / bird are deployed on the ml-serve-ctrl nodes, but the istio webhook svc seems not reac... [17:50:45] * elukey afk! [18:02:08] oof trying to rebase the editquality repo to get the most recent models....its taking foooreverrr [18:02:21] will be nice to get rid of git lfs soon [18:13:17] 10Lift-Wing, 10ORES, 10drafttopic-modeling, 10Machine-Learning-Team (Active Tasks): Create a KFServing model server for drafttopic models - https://phabricator.wikimedia.org/T285883 (10ACraze) @kevinbazira I uploaded the enwiki articletopic model to our public s3 bucket, here is the storage uri: s3://wmf-... [18:19:21] It really will. Also then I won't get the occasional ping from folks diplomatically asking if we still need LFS [18:50:17] 10AI-Governance, 10Lift-Wing: Investigate storing model metadata on Wikidata - https://phabricator.wikimedia.org/T286508 (10ACraze) [18:51:06] 10artificial-intelligence, 10Documentation, 10Machine-Learning-Team (Active Tasks): Experiment with on-wiki model documentation - https://phabricator.wikimedia.org/T276398 (10ACraze) I experimented with the [[ https://github.com/tensorflow/model-card-toolkit | model-card-toolkit ]] module a couple weeks ago... [18:51:44] 10AI-Governance, 10Lift-Wing: Investigate storing model metadata on Wikidata - https://phabricator.wikimedia.org/T286508 (10ACraze) [18:51:46] 10Lift-Wing, 10Machine-Learning-Team, 10ORES, 10artificial-intelligence, 10Documentation: Model Reporting - https://phabricator.wikimedia.org/T276397 (10ACraze) [20:10:25] 10AI-Governance, 10Lift-Wing: Investigate storing model metadata on Wikidata - https://phabricator.wikimedia.org/T286508 (10calbon) Tajh brought up a good point about what happens when someone edits the auto-generated data. Does the script just rewrite it? [20:49:39] 10AI-Governance, 10Lift-Wing: Investigate storing model metadata on Wikidata - https://phabricator.wikimedia.org/T286508 (10ACraze) That's a good point, my guess would be that we could just rewrite the data using the script's most recent revision and then merging any new data from training etc.. [20:50:21] chrisalbon: do let me know what you think? I'm intrigued about copilot [22:02:17] 10AI-Governance, 10Lift-Wing, 10Machine-Learning-Team (Active Tasks): Investigate storing model metadata on Wikidata - https://phabricator.wikimedia.org/T286508 (10ACraze)