This week is about orchestration of machine learning projects with Mage AI with integration of MLFLow
Mage is an open-source, hybrid framework for transforming and integrating data.
Link to practice repo - MLOps Orchestration with Mage
To start mage ai,
docker pull mageai/mageai:latest
docker run -it -p 6789:6789 -v $(pwd):/home/src mageai/mageai /app/run_app.sh mage start [project_name]
For my project, I used this repo with docker compose template.
git clone https://github.com/mage-ai/mlops.git
cd mlops
./scripts/start.sh
Here is my project. MLOps Orchestration with MageAI
To start my project, clone this repo and
cd mlops
./scripts/start.sh
Here is my homework link
In my homework, I built pipeline with four blocks
- Ingest (data loader)
- Data Preparation (transformer)
- Train Model (transformer)
- Log the model with MLFlow (data exporter)
The trained model is registerd and kept in artifact and the feature matrix is also recorded in artifact with MLFlow.