This month you will find:
- 🛠 New Tutorials and Guides,
- 🤫 VS Code extension,
- 📖 Doc Updates!,
- 🎥 August Meetup Video available,
- 🚀 and more!
- Jeny De Figueiredo
- September 14, 2021 • 4 min read
Welcome to September! We'll kick off this month's Community picks with a four-part series by Tezan Sahu on the Fundamentals of MLOps. Tehan introduces readers to the core ideas behind taking the best practices of DevOps and how they are being adapted to machine learning projects that deploy large scale AI powered applications. The series includes:
- Part 1: A Gentle Introduction to MLOps
- Part 2: Data & Model Management with DVC We love this part best! ❤️😉
- Part 3: MLExperimentation with PyCaret
- Part 4: Tracking with MLFlow & Deployment with Fast API
Tezan Sahu's 4 part series on the Fundamentals of MLOps Source link
If you follow the steps through this series, you will learn how to build and deploy an end-to-end ML project - all the steps leading to production!
This month Miguel Méndez of Gradiant brings us a guide on object detection using the MMdetection framework in conjunction with DVC to design the pipeline, version models and monitor training progress. This follows his first guide covering how to version your datasets with DVC, which we shared in the July Heartbeat.
In this new guide, you'll gain a thorough understanding of the steps, have access to his repo for the project, and find his thoughts on scaling hyperparameter tuning through this open issue about exeperiments that we are trying to resolve. Join the conversation! We'd love your input!
Miguel Méndez' second guide in a series using DVC in an object detecton project Source link
It was just a few short months ago when Hrittik Roy joined us at his first DVC Office Hours. Now he's written DVC (Git for Data): A Complete Tutorial on DVC and how it solves the challenges of ML engineers. In this piece he takes you through set up, pipeline and versioning, experiments and sharing through our built in shared caching, so that you and your teammates can reduce resource use when focusing on a subset of datasets as you move through your project.
Hrittik Roy's Complete Intro on DVC Source link
In case you missed it, Andy Kurenkov tweeted that he finally got around to writing about his list of 21 favorite AI Newsletters. You can find the article right here. Be sure to check it out and get reading…
One PhD student’s curated list of 21 newsletters to help you keep up with AI news and research
We know there were a lot of peeps out on holiday over the last month so let me fill you in!
Paige Bailey let the cat out of the bag with her tweet about the developent of our VS Code extension for DVC. We're getting closer every day! If you'd like to be a part of the beta testing (how could you not?) join us here.
Paige Bailey let's the cat out of the bag Source link
As promised, we will be adding this section to the Heartbeat each month so that you can stay in the know about the doc updates that will most impact your workflows. You won't want to miss these…
First up, a new doc on our Fast and Secure Data Caching Hub. Checkout this doc to learn how DVC's built-in data caching lets you implement a simple and efficient storage layer globally - FOR YOUR ENTIRE TEAM. This lets you:
- ⏱ Speed data transfers from massive object stores currently on the cloud
- 💰 Pay only for fast access to frequently-used data
- 🙅🏻♂️ Avoid extra downloads and duplicating data
- ⚡️ Switch data inputs fast (without re-downloading) on a shared server used for machine learning experiments.
Status: Must read. 📖
Fast and Secure Data Cachin Hub Source link
Is this your life?
Is this your life? Source link
Our latest doc, Continuous Integration and Deployment for Machine Learning, shows you how to move from the above chaos to CI/CD victory through:
- ✅ Data validation
- ✅ Model validation
- 🎟 Provisioning
- 📈 Metrics
Read the whole doc to learn how DVC and CML will enable you to run entire experiments/research online and remove most of your managment headaches to look more like this. 👇🏼
Traditional ML meets CI/CD with DVC and CML Source link
Cleaning Up Experiments has been made bright and shiny and new to do the same with your experiments. Be sure to check it out!
This Thursday at our September Office Hours Meetup, Milicia McGregor will be presenting her tutorial on Using Experiments For Transfer Learning. Join us on September 16th at 3:00 pm UTC! RSVP at this link below! 👇🏼
DVC Office Hours - Using Experiments For Transfer Learning
We'll be introducing some new team member next month, but we are still hiring. So do checkout our open positions here to find details of all the positions including:
- Senior Front-End Engineer (TypeScript, Node, React)
- Senior Software Engineer (ML, Dev Tools, Python)
- Senior Software Engineer (ML, Data Infra, GoLang)
- Machine Learning Engineer/Field Data Scientist
- Developer Advocate (ML)
- Director/VP of Engineering (ML, DevTools)
- Director/VP of Product (ML, Data Infra, SaaS)
- Director/VP of Operations/Chief of Staff
Please pass this info on to anyone you know that may fit the bill. We look forward to new team members! 🎉
Last week this Tweet brought us another 300 Twitter followers, catapulting us over 3000! Thanks Community for joining us on this MLOps ride! More to come! 🚀
Startups I'm *incredibly* bullish about: @Stripe, @IterativeAI, @HuggingFace, and @Explosion_AI.— 👩💻 Paige Bailey (@DynamicWebPaige) September 7, 2021
If you're an engineer/PM considering a career change (and it's that time of the year again, no? 😆)—but want to opt away from FAAMG, definitely consider one of the companies above.
Do you have any use case questions or need support? Join us in Discord!
Head to the DVC Forum to discuss your ideas and best practices.