Hello, Question, what is the recommended way to do model serving and model registry with Flyte? I understand Flyte primarily concerns itself with orchestration.

Hi, I have a questions with regards to cluster-resource-attribute. I assume the sync to K8s itself happens periodically, what is the period/delay post an update?

Hello Can we install ray in sandbox with pip? It seems, ray don’t have right version for sandbox ((

Hello, flyte team. I'd like to ask union.ai, especially union cloud. I have submitted demo usage application in union.ai to check two things. 1. when research engineers process ML Work, there have been huge I/O process in on-premise server. My team has used A100 & V100 nvidia server, but these cannot handle all the I/O process during ML workflow. Does Union cloud provide enough IOPS? 2. How much cost I spend if I am using union cloud? I have not seen any price list on the website 3. when I can use union cloud after I submitted the demo application? Thank you!

Team, i am trying to update the spark logging link with below template uri.

- spark.hadoop.fs.AbstractFileSystem.s3a.impl: "org.apache.hadoop.fs.s3a.S3A"
          - spark.hadoop.fs.s3a.multipart.threshold: "536870912"
          - spark.blacklist.enabled: "true"
          - spark.blacklist.timeout: "5m"
          - spark.task.maxfailures: "8"
        logs:
          mixed:
            kubernetes-enabled: true
            kubernetes-template-uri: '<https://app.datadoghq.com/logs?query=kube_namespace:{{> .namespace }} pod_name:{{ .podName }}&cols=&index=&messageDisplay=inline&stream_sort=time,desc&viz=stream&from_ts=1663061891694&to_ts=1663148291694&live=true'
          system:
            kubernetes-enabled: true
            kubernetes-template-uri: '<https://app.datadoghq.com/logs?query=kube_namespace:{{> .namespace }} pod_name:{{ .podName }}&cols=&index=&messageDisplay=inline&stream_sort=time,desc&viz=stream&from_ts=1663061891694&to_ts=1663148291694&live=true'
          all-user:
            kubernetes-enabled: true
            kubernetes-template-uri: '<https://app.datadoghq.com/logs?query=kube_namespace:{{> .namespace }} pod_name:{{ .podName }}&cols=&index=&messageDisplay=inline&stream_sort=time,desc&viz=stream&from_ts=1663061891694&to_ts=1663148291694&live=true'

however flytepropeller pod is not getting created and throwing below error. time="2022-09-14T112612Z" level=error msg="\nWhile parsing config: yaml: line 6: mapping values are not allowed in this context"

Hey all! 👋 We’ll use this channel for announcements from now on. Posting access here is being restricted as we want to emphasize Flyte announcements. Please redirect your questions to #ask-the-community and we’d be more than happy to help! We also have a couple more channels that you can join. Pasting an excerpt from our onboarding message for your reference. 👋 Head over to #introductions to introduce yourself. We’d love to know about you and what brings you here. 🎤 #announcements is used for workspace-wide communication and announcements. Keep your tabs on it! ❓ Have Flyte questions? #ask-the-community. 📦 Have questions regarding flytekit, our Python/Java SDK? #flytekit is the place to go. 🚀 Planning to deploy Flyte? Hit us up in #flyte-deployment. 💡 Have suggestions or ideas on how to make Flyte better? Jump to #feature-discussions. 🎤 We often host events and talk about Flyte at meetups and conferences. Join #events to stay tuned. 🤖 Want to stay updated on UnionML — a wrapper we built to simplify building and deploying ML microservices? Join #ecosystem-unionml. 🌏 For non-work banter and water cooler conversations, jump to #random. We’re also working on a Slack etiquette guide to make sure everyone goes by the rules. Happy slacking!

Hi everyone, Are you an ML Developer or a Data Engineer who wants a fast start using Flyte? This cheat sheet offers an easy-to-browse set of code snippets to speed up your mastery of Flytekit syntax and features. Check out the blog for more information and download instructions. https://blog.flyte.org/flyte-cheat-sheet

📣 📣 Seattle peeps & anyone in town! Join @Eduardo Apolinario (eapolinario) from Union.ai and @Richard Liaw from Anyscale in person tomorrow at 5pm for a machine learning hangout! Register on Meetup - only 10 spots left! Hyatt Regency Bellevue, Wed Sept 21 @5pm PST https://www.meetup.com/seattle-data-machine-learning-and-infrastructure/events/288290270/

Machine learning orchestration is not a mature discipline yet. Building ML pipelines per se is a complex process, which is why we developed new ML add-ons to take the load off your shoulders: • Flyte Decks 📊 is Flyte’s way of visualizing data, such as outputs and intermediary data. • The Flyte PyTorch types enables automatic conversion of tensors from GPU ➡️ CPU. • Bridge the gap between developing and deploying ML models with Flyte ONNX types for ScikitLearn, PyTorch, and TensorFlow. • Spark ✨ Pipelines can now be passed seamlessly among the Flyte tasks. • Data and model monitoring are important phases of ML pipelines. They help gauge how efficient your data and ML models are. Use WhyLab’s whylogs 🪵 from within Flyte to monitor your data and models. More details in @Samhita Alla's latest blog How We Are Simplifying the Orchestration of Machine Learning Pipelines.

office hours now for those interested:

Zoom link:

https://us06web.zoom.us/j/88174742746?pwd=YkZDRVFuU0F6eXNPU0xVYXh0R1l2Zz09

Meeting ID: 881 7474 2746

Passcode: 390635

Thanks to everyone who attended the Flyte Community Sync event yesterday! Take a look at some , get the scoop on , and learn about from Data Scientist, @krishna Yerramsetty. Join us again on Oct 4th! Meeting notes & slides / Calendar Invite

Come join the weekly office hours! I’ll be on Zoom for the next 30 mins to answer your Flyte questions.

📣 Contributors of the Month 📣 September has been a UnionML 🤖 and DBT dbt month, thanks to @Zev Isert who worked on a UnionML PR that checks and warns about dirty Git index during deployment and Arief Rahmansyah for the DBT plugin! Dear Flyte x DBT users, We heard you! @Arief Rahmansyah and @Eduardo Apolinario (eapolinario) added a Flytekit plugin for DBT. It should be available in the upcoming Flytekit release; please keep tabs on the #announcements channel! With love, flyte All contributors of the month are available on flyte.org.

Deciding between Data, Machine Learning, or even Infrastructure orchestration? 🤔😵‍💫🤯 Dive into each orchestration technique in the blog below for an informed decision about which works best for your use case. 💡👌🙂 https://www.union.ai/blog-post/orchestration-for-data-machine-learning-and-infrastructure

Come join the weekly office hours! I’ll be on Zoom for the next 30 mins to answer your Flyte questions.

hi all! 👋 Join us for the UnionML OSS Planning 📋🛣️ today at 12pm EST, 9am PST! Today we’ll: • go over the revised roadmap • cover UnionML hacktoberfest • demo: patch deployment 💻 Zoom link: https://us06web.zoom.us/j/85830795287?pwd=K0tXK3l2bzE2UEFYbUhGZnNRRWNKZz09 🗓️ Calendar event: https://app.addevent.com/event/tj14110550/

UnionML OSS Planning starting now! https://us06web.zoom.us/j/85830795287?pwd=K0tXK3l2bzE2UEFYbUhGZnNRRWNKZz09

Come join for weekly office hours now: https://us06web.zoom.us/j/88174742746?pwd=YkZDRVFuU0F6eXNPU0xVYXh0R1l2Zz09

Come join the weekly office hours! I’ll be on Zoom for the next 30 mins to answer your Flyte questions.

Flyte, as well as UnionML - a machine learning wrapper built on top of Flyte, #ecosystem-unionml - are participating in Hacktoberfest 2022! Here are 3 reasons why you should contribute, how to register/contribute/earn prizes, plus common FAQs. https://blog.flyte.org/meet-flyte-and-unionml-at-hacktoberfest-2022

Flyte Monthly Issue #12 is out! Read up on Hacktoberfest, Flyte's Ray integration, Tweet of the week, and so much more! Subscribe here.

📣 📣 📣 The first Flyte Quarterly Planning meeting open to the community will be on Mon, Oct 3. Join the planning process, increase community involvement, and surface high-level feature requests to help drive the roadmap going forward. Join us at 11:30am PT. Calendar Invite & Zoom Link

🕦 The first Flyte Quarterly Planning meeting kicks off in about one hour! 🕦

Also, our community sync is right around the corner! - Take a look into orchestration tools, and how Flyte is simplifying orchestration - Watch a UnionML patch deployment demo (check out #ecosystem-unionml) - And hear the latest on the state of Flyte v1.2.0, plus plans for the next release. Tuesday Oct 4, 9am PT

Hey all! We just published an article on why Woven Planet switched some of their Spark workflows to Flyte map tasks. Thanks @Vijay Saravana, @Alex Bain and @varsha Parthasarathy for writing this fantastic piece! https://blog.flyte.org/flyte-map-tasks-a-simpler-alternative-to-apache-spark

If you missed Flyte's Quarterly Planning Meeting, check out the , and mark your calendars for the next one in early January. In the meantime, send questions, comments or suggestions to @Eduardo Apolinario (eapolinario), or create an issue on GitHub!

👋 Come join the weekly office hours! I’ll be on Zoom for the next 30 mins to answer your Flyte questions.

🎉 Flyte v1.2.0 is out! 🎉 Complete with: - Ray integration (reducing workflow Kubernetes resources - check out the Ray & Flyte blog post) - Perf improvements - DBT & hugging face Flytekit plugins, and - PyFlyte Run improvements Check out the full release notes at https://github.com/flyteorg/flyte/releases/tag/v1.2.0

hi all come join for weekly office hours: https://us06web.zoom.us/j/88174742746?pwd=YkZDRVFuU0F6eXNPU0xVYXh0R1l2Zz09 🖖

Hi all! 👋 Super excited to announce that Pandera has officially joined the Union.ai open source ecosystem! Pandera is the open source framework for data testing, built for data scientists and ML engineers. It enables you to define schemas for your dataframe-like objects so you can be confident in the quality of your data as it passes through your (Flyte) pipelines Please feel free to ping me if you have any questions, and give Pandera some love on product hunt! https://www.producthunt.com/posts/pandera