responsible-ai-toolbox/rai_core_flask

rai_core_flask

Provide environment detection for enabling Flask server without configuring at instantiation. Environments are implemented as classes that will either return None if the environment check fails, or return an object with all the appropriate environment settings if the check passes. The flask service can then use it to build out urls, attach required headers, and render inline visualizations with the correct methods.

Installation

pip install -e .

Usage

from rai_core_flask import FlaskHelper flask_service = FlaskHelper(ip=<ip-to-listen-on>, port=<port-to-listen-on>)

If ip is not specified then it listens on localhost by default. If port is not specified then it finds an open port in the range 8704 to 8993 and listens on it.

To register a function to listen on an route:

@flask_service.app.route("/your-api-route/<int:param_name>", methods=["GET", "POST"])
def my_func(param_name):
    return something

Release Process

1. Increment the version number in setup.py
2. Activate your virtual environment.
3. Run the release.sh script.