nbox-specimens

Welcome to nbox-specimens, your one-stop-shop for nbox examples and follow-along materials. If you're new to nbox, don't worry, we've got you covered. Here's how to get started:

pip install nbox

# Just call this once to authenticate
nbx why

If anything goes wrong, blame it on the rain or update nbox to the latest version.

🍇 How to use NBX-Projects

Our NBX-Projects tool is the Swiss Army knife of MLOps project management. It's so easy to use, even your grandparents can train a PyTorch model in no time! Check out these examples:

• projects_01: Train a PyTorch model, visualize logs, and store artifacts using NimbleBox projects.
• The stable-diffusion AI startup: [GPU] Train a Stable Diffusion model and serve it and get ready to conquer the world.
• Mastering Blackjack with Q-learning: Train a Q-learning agent to play Blackjack like a pro.

🌳 Service-Specific

We're not just a one-trick pony. Check out these projects that demonstrate how to run different kinds of workloads:

• [Serving] FastAPI: Deploy a FastAPI server on an NBX-Serving and feel like a superhero.
• [Jobs] Sklearn Tests: Train a simple k-means model and store the resulting chart in a Relic.
• [Serving] Pose Model: Deploy a mediapipe pose detection model on a Kubernetes cluster (w/o YAML 😛).
• [Jobs] DeepSpeed: [GPU] Train a MobileViT using DeepSpeed and become a Deep Learning wizard.

🌌 More ambitious projects

For those who crave a challenge, these advanced projects will test your mettle:

• Compute Fabric: Attach powerful cloud compute and supercharge your local development like a Jedi.
• Did you know that there are some really bizarre and fascinating phenomena in the world that can be explained by simple mathematical models? Dive into our 'Did you know' section and explore some of the weird and wonderful things that happen in our universe!

May the force (and nbox) be with you!

Join the Party

Want to join the party? We are always on the lookout for fresh brains! If you have an idea for a new example that you think would make the world a better place, don't hesitate to open an issue. We'll get back to you as soon as possible, preferably before the next Netflix series comes out.