A toy project for MLOps using python, pytorch, W&B, fly.io
This repository contains the implementation of various software to support end-to-end MLOps from development to deployment.
Github: Code Repository CI/CD: Github Action Container: Docker Tracking: Weight and Biases Registy: Weight and Biases Deployment: Fly.io
The use case of this repository is handwritten character recognition. This may change the future to create more challenging process of the development or support multiple projects at once.
├── main <- Release version
├── dev <- Model development
├── feat <- Using this as prefix feat/* to develop specific feature
The directory structure of your new project looks like this:
├── LICENSE
├── README.md <- The top-level README for developers using this project.
├── data
│ ├── processed <- The final, canonical data sets for modeling.
│ └── raw <- The original, immutable data dump.
├── notebooks <- notebook development.
├── models <- weights and dependencies.
├── src <- scripts for training and evaluating.
pip install -r requirements