Deploying a Machine Learning Model on Heroku with FastAPI

• Origin
• Author: Heber Trujillo heber.trj.urt@gmail.com
• Date of last README.md update: 23.04.2022

Repo Overview

This repository shows how to Deploy an end-to-end machine learning (ML) pipeline using FastAPI and Heroku.

See API documentation

The following directional graph shows the implemented ML-pipeline:

Which contains the following components:

• prepare: preprocess the census data.
• segregate: Segregates the data into test and training sets.
• train: Trains a classification inference artifact.
• evaluate: Test the fitted inference artifact.

How to Run Scripts

Dependencies Installation

1. Create and activate a virtual environment for the project. For example:

   python3 -m venv ./.venv
   source ./.venv/bin/activate

2. Install Poetry, the tool used for dependency management. To install it, run from a terminal:

   pip install poetry

3. From the virtual environment, install the required dependencies with:

   poetry install --no-root

Execute Pipeline

The ML-pipeline can be executed:

dvc repro

Run API Locally

Type the following command to run the live server locally:

 uvicorn main:app --reload

Interactive API docs

Now go to http://127.0.0.1:8000/docs

You will see the automatic interactive API documentation:

API

Documentation

The API has two endpoints, as shown in the API documentation.

GET - Root

The root domain contains a greeting and a helpful link where you can check the model card:

POST - Inference

This endpoint performs inference on new data: