/ml-serving-showcase

Implementation of an ML Model Serving with Flask, the model is LGBM trained on Kaggle titanic data.

Primary LanguageJupyter NotebookMIT LicenseMIT

ML-Serving-Showcase

License Python Version Code style: black Binder

Implementation of a ML Model Serving with Flask, the model is LGBM trained on Kaggle titanic data. To run the service:

python ml_serving.py

To test the service go to the following URL: http://127.0.0.1:5000/predict

Here is an input for example (JSON format):

[
    {"Age": 85, "Sex": "male", "Embarked": "S", "Pclass": 3},
    {"Age": 63, "Sex": "female", "Embarked": "C", "Pclass": 2},
    {"Age": 15, "Sex": "male", "Embarked": "Q", "Pclass": 1}
]

Here is an example of the API on POSTMAN: alt text