The entire model creation and deployment is explained in details in this notebook.
It is as example for creating a classification model for production
We will walk through:
- building a Deep Neural Network that predicts labels given features (using scikit-learn and Keras)
- build a REST API that predicts labels based on the model (using Flask and gunicorn)
- deploy the model to production on Google App Engine
Requirements:
- Python 3.7
- Google Cloud Engine account
- Google Cloud SDK
Clone this repository:
git clone [email protected]:nobari/deep.git
cd deep
Install libraries:
pip install -r requirements.txt
flask run
curl -d '{"0": "Brooklyn", "1": 123, "2": -123, "3": 123, "4": 123}' -H "Content-Type: application/json" -X POST http://localhost:5000
gcloud app deploy