An Microservice ML Endpoint based on AutoML Solution using Ludwig that uses Continuous Integration & Continuous Delivery
This project showcases an Automated Machine Learning (AutoML) solution using Ludwig, Uber's open source deep learning toolbox. The advantages of using Ludwig are listed under the Ludwig Features section below. The project also showcases a microservice architecture that uses Continuous Integration & Continuous Delivery (CI/CD) to deploy the model as a Flask API endpoint. The CI/CD pipeline is implemented using GitHub Actions. The project also includes a frontend that allows users to interact with the model via a web interface using Streamlit in the Usage section.
- Easy-to-Use: Ludwig is designed to be user-friendly and requires no programming experience. All you need to train a model and make predictions is a tabular dataset and a few lines of command-line code.
- Flexible: Ludwig enables you to train a deep learning model using various types of input data such as text, images, and more. You can also combine different types of input data to create a hybrid model.
- Versatile: Ludwig is suitable for a wide range of tasks such as image classification, text classification, and time series forecasting.
- Python 3.7+
- See reqirements.txt for more details
Clone the repository to your local machine:
Install the dependencies:
pip install -r requirements.txtLudwig requires the data to be in a tabular format such as CSV. The column names of the CSV file will be used as the feature names.
- Model Training (Optional):
python train_automl.py --data-path ./data/agnews.csv --config-path ./config/config.json- Model Generation (Optional):
make gen- Model prediction using pre-trained model:
python predict_automl_torch.py --predict-path ./data/text_to_predict.csv- API server:
flask run --host=0.0.0.0 --port=80- Or run with Docker:
docker run -p 8080:8080 szheng3/automl-python- Frontend:
streamlit run frontend.pyThis project makes use of the fantastic Ludwig toolbox developed by Uber's AI team. We thank them for their work and for making it open source.