Machine Learning Microservice with FastAPI

This application serves machine learning models trained on the Titanic dataset, allowing API-based predictions and model evaluation.

API Overview

There are two versions of the API (/v1 and /v2), each offering two endpoints: /predict and /score.

• /v1: Utilizes a regularized logistic regression model.
• /v2: Utilizes a random forest model.
• /predict: Returns the predicted probability of passenger survival.
• /score: Provides model metrics on a test set (ROC AUC, accuracy, and recall).

Deployment

The app is deployed with continuous integration to AWS ECS using a Terraform template. GitHub Actions handles automated deployments on changes to the main branch.

Local Usage

To run the app locally using Docker:

$ git clone https://github.com/ftrifoglio/fastapi-demo.git
$ cd fastapi-demo
$ docker-compose up

Roadmap

• Add unit tests to titanic_model and api_utils