2024-02-20 19:49:27 -08:00
|
|
|
|
2024-03-02 09:11:52 -08:00
|
|
|
#MLOps
|
|
|
|
|
|
|
|
|
|
I have an ML lab environment at home that I'm converting to something with better MLOps standards
|
|
|
|
|
|
|
|
|
|
## WORK IN PROGRESS
|
|
|
|
|
|
|
|
|
|
Insert tacky animated gif of a construction worker here
|
|
|
|
|
|
|
|
|
|
## Goals
|
2024-02-20 19:49:27 -08:00
|
|
|
|
2024-02-20 20:09:32 -08:00
|
|
|
Broadly, the goals are:
|
|
|
|
|
- Local dev environment that's as similar as possible to production
|
|
|
|
|
- A managed Jupyter Lab environment for experimentation
|
|
|
|
|
- Do something sensible with features and feature extraction
|
|
|
|
|
- A good mockup of an operationalized model
|
|
|
|
|
- CI/CD pipeline with local access to test/verification
|
|
|
|
|
|
|
|
|
|
|
2024-02-20 19:49:27 -08:00
|
|
|
## Requirements
|
|
|
|
|
|
|
|
|
|
- docker
|
|
|
|
|
- https://github.com/NVIDIA/nvidia-container-toolkit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|