/ml-pipeline

A simple ML pipeline project. Nothing too fancy.

Primary LanguagePython

Machine Learning Pipeline

An End-to-End Machine Learning pipeline deployed as a REST API.
API developed with FastAPI

Data

Medical Insurance Payout
https://www.kaggle.com/datasets/harshsingh2209/medical-insurance-payout

Process:

  • POST training data on train/
  • Data gets preprocessed and split
  • Model pipeline is created
  • Random Forest Regressor model gets trained
  • Evaluated
  • POST test sample on test/

Setup:

  1. Clone the repository
  2. Create virtualenv and run pip install -r requirements.txt
  3. To launch server, run uvicorn app:app --reload
  4. Test with Python requests or Postman