Simplify your fitting and deployment routine today!
| Feature | Description | |
|---|---|---|
| 🔄 | Flexible Training & Validation Options | Supports multiple methods for training and validating models |
| 🛠️ | Hyperparameter Optimization | Includes built-in hyperparameter tuning with options for Grid Search, Bayesian optimization, and Optuna optimization |
| 🎻 | Model ensembling | Provides various methods for combining predictions from multiple models to enhance accuracy, stability, and generalization.. |
| ⚙️ | Easy Integration | Quickly get started with a new machine learning project by cloning or forking Coginets, you just need to defne a custom pipeline for your dataset. |
| 🌐 | Model Deployment | Offers model and pipeline serialization and integrates automatically with FastAPI for serving models predictions as an API endpoint. |
| 🐳 | Containerization | Includes Dockerfile and Docker Compose configuration for straightforward deployment in any environment. |
| 🤖 | CI/CD Ready | GitHub Action templates included:
|
| ✅ | Unit Testing | Ensures code reliability and correctness through automated unit tests. Adding tests for new modules is straightforward—just compile a configuration class, and they will be auto-generated. |
| 📏 | Code Quality | Maintains high code quality standards through SonarQube quality gate. |
A complete example is provided in the repository, showcasing Coginets with a home price prediction model. This example includes:
The dataset for training A Docker configuration for easy deployment
Check out the live preview on Render and try out inference.
Note: The live preview will cold-start as you enter the link, it could take up to 1min to fully load.
Access the API: After installing with the following methods, navigate to http://localhost:8080/docs or http://localhost:8080/redoc to explore the
interactive API documentation and start making predictions!
- Pull the Docker Image:
docker pull ghcr.io/manuel-materazzo/coginets-example:latest
- Run the Container:
docker run -d -p 8080:80 manuel-materazzo/coginets-example
- Run docker compose:
docker-compose up
- Clone Coginets repository:
git clone https://github.com/Manuel-Materazzo/Coginets.git cd Coginets - Install the required dependencies:
pip install -r requirements.txt
This project is licensed under the MIT License. See the LICENSE file for more details.