This repository contains my guide to deploy ML models. The repository was forked from
deploying-machine-learning-models
which contains the companion code of the Udemy course Deployment of Machine Learning Models by Soledad Galli & Christopher Samiullah.
The guide or notes for my future self done after following the course are in: ./ML_Deployment_Guide.md
.
- Presentations: provided as a Dropbox download link, located in
./udemy_ml_deployment/deployment_of_ML_presentations
; but not committed. - Datasets: downloaded from kaggle: House Prices - Advanced Regression Techniques; located in
./data/house-prices-advanced-regression-techniques
; but not committed.
- Overview of Model Deployment
- Machine Learning System Architecture
- Research Environment: Developing a Machine Learning Model
- Packaging the Model for Production
- Serving and Deploying the Model via REST API - FastAPI
- Continuous Integration and Deployment Pipelines - CicleCI
- Deploying the ML API with Containers
- Differential Tests
- Deploying to IaaS (AWS ECS)
Continue in ./ML_Deployment_Guide.md
for the detailed guide/notes.
Also check these links:
- Through deployment example: census_model_deployment_fastapi.
- My personal notes on the Udacity MLOps nanodegree: mlops_udacity. The module guide about deployment:
MLOpsND_Deployment.md
- Check this example deployment: census_model_deployment_fastapi
- Check my personal notes on the Udacity MLOps nanodegree: mlops_udacity:
- The module guide about deployment:
MLOpsND_Deployment.md
- Example and exercise repository related to the topic of deployment: mlops-udacity-deployment-demos.
- The module guide about deployment:
- My guide on CI/DC: cicd_guide
- My boilerplate for reproducible ML pipelines using MLflow and Weights & Biases: music_genre_classification.
- A very simple Heroku deployment with the Iris dataset and using Flask as API engine.
- Notes on how to transform research code into production-level packages: customer_churn_production.
- My summary of data processing and modeling techniques: eda_fe_summary.
Notes by Mikel Sagardia, 2022.
No guarantees.