iris-fast-api

This is an example to explain how to serve a ML model as an API by using FastAPI

Serving Iris Classifier with FastAPI

Data

The Iris dataset is a simple, yet popular dataset consisting of 150 observations. Each observation captures the sepal length, sepal width, petal length, petal width of an iris (all in cm) and the corresponding iris subclass (one of setosa, versicolor, virginica).

Setup

• Create a virtual environment with:

python3 -m venv venv

• Activate the virtual environment

source venv/bin/activate

Install the other libraries

Run the following command to install the other libraries.

pip install -r requirements.txt

Usage

Run FastAPI

Run next command to start the api locally

uvicorn iris.v1.app:app --port 5001 --reload
uvicorn iris.v2.app:app --port 5002 --reload

Test request

The input is a JSON with the following fields:

• sepal_l
• sepal_w
• petal_l
• petal_w

Corresponding values are the measurements in cm.

Example request:

curl 'http://localhost:8080/iris/classify_iris' -X POST -H 'Content-Type: application/json' -d '{"sepal_l": 5, "sepal_w": 2, "petal_l": 3, "petal_w": 4}'