Metaflow Tags: Tags in CI/CD
Key Takeaways
Metaflow tags are used in conjunction with CI/CD systems like GitHub actions to streamline MLOps workflows
Full Transcript
let's turn into more production oriented use cases for tagging so now imagine that you're a data scientist who has been iterating on a workflow locally let's say we execute this workflow we keep making changes um whatever it might be and and at some point like we are pretty confident that the workflow does what we wanted it to do so uh of course it is it is definitely considered a good practice that you used git for version control and not only that but maybe you have some kind of a ci cd setup you used to create pull request and so forth so we can do the usual thing that we do when developing software so let's say that we have been working on on some feature brands so this is new awesome model branch and now we are really happy with our results um so we uh of course like first add our changes there and like we commit uh add new awesome model right like this right and um now we push our branch to github and um as usual what we want to do here by the way i can show you that we have all these past executions that have been run locally first but now when we go to github we are ready to create a pull request and we can say that okay we are adding new awesome model here we create the pull request and what i did here is that i we have also uh github actions set up and what the github actions do i can show you here is that they first execute the workflow here they execute a test run like making sure that the flow executes correctly of course this is your opportunity to set up any kind of a test you want to have in place oftentimes we have unit tests in place the test individual modules um that are that are like kind of the part of your workflow maybe you have an end to end integration test like for the whole workflow like what we are doing here but the key part here is that like after the test is successful we will tag the or the github actions will tag the the successful run so you are able to see uh like when when the kind of the the ci cd tests have passed successfully and then you can take further action based on that information for instance you could deploy the model automatically you could deploy the workflow automatically whenever that happens so now we can see that the tests executed successfully and after the test executed successfully we have an action to tag the corresponding executions so now if we go back to the to the meta flow ui here we see that there is a one more execution of the same workflow but in this case like we know that it was executed by github actions we just tag it accordingly and then like we have the tag mentioning that okay so we know that the tests were okay and now this is quite convenient because now in this case like you see all your local executions and then the github executions mixed up but i mean for instance you could imagine like having this view like where you only see like all the all the successful uh cicd execution so you can keep track of all the pushes all the model versions that uh resulted due to kind of a positive tests like successful tests being executed by github actions
Original Description
Learn how to use Metaflow tags in conjunction with CI/CD systems like GitHub actions
Find out more about how we think about MLOps, OSS, and human-centric data science tools here: https://outerbounds.com/
Watch on YouTube ↗
(saves to browser)
Sign in to unlock AI tutor explanation · ⚡30
Playlist
Playlist UU5h8Ji6Lm1RyAZopnCpDq7Q · Outerbounds · 9 of 60
1
2
3
4
5
6
7
8
▶
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Metaflow GUI for monitoring machine learning workflows
Outerbounds
Metaflow Cards [no sound]
Outerbounds
Fireside chat #1: How to Produce Sustainable Business Value with Machine Learning
Outerbounds
Fireside chat #2: MadeWithML.com -- Teaching Practical Machine Learning
Outerbounds
Metaflow on Kubernetes and Argo Workflows [no sound]
Outerbounds
Fireside chat #3: Reasonable Scale Machine Learning -- You're not Google and it's totally OK
Outerbounds
Metaflow Tags: Programmatic Tagging
Outerbounds
Metaflow Tags: Basic Tagging
Outerbounds
Metaflow Tags: Tags in CI/CD
Outerbounds
Metaflow Tags: Tags and Namespaces
Outerbounds
Metaflow Tags: Tags and Continuous Training
Outerbounds
Fireside chat #4: Machine Learning and User Experience -- Building ML Products for People
Outerbounds
Fireside Chat #5: Machine Learning + Infrastructure for Humans
Outerbounds
Metaflow Sandbox Demo: Free Data Science Infrastructure In the Browser
Outerbounds
Metaflow on Azure
Outerbounds
Fireside Chat #6: Operationalizing ML -- Patterns and Pain Points from MLOps Practitioners
Outerbounds
ML engineering vs traditional software engineering: similarities and differences
Outerbounds
Why data scientists love and hate notebooks: velocity and validation
Outerbounds
What even is a 10x ML engineer?
Outerbounds
The 4 main tasks in the production ML lifecycle
Outerbounds
Is the premise of data-centric AI flawed?
Outerbounds
The 3 factors that Determine the success of ML projects
Outerbounds
Fireside Chat #7: How to Build an Enterprise Machine Learning Platform from Scratch
Outerbounds
Run Metaflow on any cloud: Google Cloud, Azure, or AWS [no sound]
Outerbounds
Metaflow on GCP
Outerbounds
Fireside Chat #8: Navigating the Full Stack of Machine Learning
Outerbounds
How to Build a Full-Stack Recommender System
Outerbounds
Modernize your Airflow deployments with Metaflow - zero-cost migration [no sound]
Outerbounds
Easy Airflow DAGs for ML and data science with Metaflow [no sound]
Outerbounds
Fireside chat #9: Language Processing: From Prototype to Production
Outerbounds
How to build end-to-end recommender systems at reasonable scale
Outerbounds
Full-Stack Machine Learning with Metaflow on CoRise
Outerbounds
Natural Language Processing meets MLOps
Outerbounds
Fireside Chat #10: Large Language Models: Beyond Proofs of Concept
Outerbounds
What even are Large Language Models?
Outerbounds
How to get started with LLMs today
Outerbounds
LLMs in production
Outerbounds
Accessing secrets securely in Metaflow [no audio]
Outerbounds
Fireside Chat #11: The Open-Source Modern Data Stack
Outerbounds
Fireside chat #12: Kubernetes for Data Scientists
Outerbounds
Behind the Screen: How Amazon Prime Video ships RecSys models 4x faster
Outerbounds
Fireside chat #13: Supply Chain Security in Machine Learning
Outerbounds
Quick Delivery, Quicker ML: DeliveryHero's Metaflow Story
Outerbounds
Crafting General Intelligence: LLM Fine-tuning with Metaflow at Adept.ai
Outerbounds
Fuelling Decisions: How DTN Powers Gas Pricing and Data Science Collaboration
Outerbounds
From Kitchen to Doorstep: Optimizing Data Science Velocity at Deliveroo
Outerbounds
Building a GenAI Ready ML Platform with Metaflow at Autodesk
Outerbounds
Media Transcoding for 10 Million users and beyond with Metaflow at Epignosis
Outerbounds
Telematics with Metaflow: How Nirvana Insurance built a large-scale Risk Estimation platform
Outerbounds
Fireside chat #14: Generative AI and Machine Learning for Film, TV, and Gaming
Outerbounds
The Past, Present, and Future of Generative AI
Outerbounds
Building Production Systems with Generative AI, Machine Learning, and Data
Outerbounds
A Custom Fine-Tuned LLM in Action (LLMs, RAG, and Fine-Tuning: An Interactive Guided Tour Part 5)
Outerbounds
Building Live Production Systems with RAG (LLMs & RAG: An Interactive Guided Tour Part 4)
Outerbounds
Better Relevancy with RAG (LLMs, RAG, and Fine-Tuning: An Interactive Guided Tour Part 3)
Outerbounds
Working with OSS LLMs (LLMs, RAG, and Fine-Tuning: An Interactive Guided Tour Part 2)
Outerbounds
Hitting OpenAI and Other Vendor APIs (LLMs, RAG, and Fine-Tuning: An Interactive Guided Tour Part 1)
Outerbounds
Production Systems with Generative AI (LLMs, RAG, & Fine-Tuning: An Interactive Guided Tour Part 0)
Outerbounds
LLMs in Practice: A Guide to Recent Trends and Techniques
Outerbounds
Metaflow for distributed high-performance computing and large-scale AI training
Outerbounds
More on: ML Pipelines
View skill →Related AI Lessons
⚡
⚡
⚡
⚡
`wrangler dev --remote` silently writes to your production KV namespace — here's the fix
Dev.to · 강해수
Qwen 3.6 27B Is the Local Dev Sweet Spot — Here's Why
Dev.to · Carter May
Deploying Spring Petclinic Microservices with Docker Compose: An End-to-End DevOps Deployment Experience
Dev.to · Nice Nwogu
Qwen 3.6 27B Is the Local Dev Sweet Spot — Here's Why
Dev.to · Carter May
🎓
Tutor Explanation
DeepCamp AI