MLopsTechnicalChallenge

This project is initially built for a technical challenge to join Junior MLOPS Postion

FastAPI Sentiment Analysis Service

It is a sentiment analysis service built with FastAPI. It utilizes a machine learning model(ROBERTA) to analyze text and determine sentiment (Positive, Negative, Neutral). The service is containerized using Docker for easy deployment and scalability.

Features

• Sentiment analysis of textual data.
• RESTful API endpoint.
• Integration with RDS(PostgreSQL) database for logging predictions.
• Docker support for easy deployment.

Getting Started

These instructions will get your copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

• Python 3.12+
• Docker
• PostgreSQL (For local testing or if not using AWS RDS)

Installation

1. Clone the repository:

   git clone https://github.com/lillianassefa/MLopsTechnicalChallenge.git
   cd MLopsTechnicalChallenge/app
   

2. Set up a virtual environment (Optional but recommended):

   python -m venv venv source venv/bin/activate # On Windows use venv\Scripts\activate

3. Install Dependencies:

   pip install -r requirements.txt

4. Set Environment Variable

   Set up the following environment variables, either in a .env file or in your environment: DB_NAME: postgres DB_USER: postgres DB_PASSWORD: Database password DB_HOST: microservicemodel.crwuq2iiy62m.us-east-1.rds.amazonaws.com DB_PORT: 5432 AWS_ACCESS_KEY_ID: AWS access key AWS_SECRET_ACCESS_KEY: AWS secret access key BUCKET_NAME: modelforpredict

5. Running the application locally

   uvicorn main:app --reload

Running with docker

1. Build docker image docker build -t microservicesentimentpredict

2. Run docker container

   docker run -d -p 8000:8000 fastapi-sentiment-app

Usage

There is one api endpoint which is /predict which is used to accept text input from user and outputs the sentiment value.

API Endpoints

GET /ping: Health check endpoint. POST /predict: Endpoint to submit text for sentiment analysis. Request body format:

{"text": "Your sample text here"}

Testing

The application can be tested using Postman or the Swagger UI at http://localhost:8000/docs.

Contributing

Please feel free to contribute to this project and give suggestions

License

This project is licensed under the MIT License - see the LICENSE.md file for details.