MLOPS Projects

---

OVerview

This repository provides a comprehensive approach to Machine Learning Operations (MLOps), integrating machine learning models into production with automation, monitoring, and scalability. It covers best practices, CI/CD pipelines, model versioning, and deployment strategies.

Projects

This repository includes multiple MLOps projects, each focusing on different aspects of machine learning model development, deployment, and monitoring. The projects are structured as follows:

Employee Attrition Prediction

    - Uses Logistic Regression for predicting employee attrition.
    - Implements Flask for web-based model interaction.
    - Features automated data preprocessing, model training, and deployment using Docker and Kubernetes.

LLM-Based Simple models using Hugging Face

    - Built simple LLM project using Hugging Face's open source models on
        - text summarization, 
        - text generation, 
        - sentiment-analysis, 
        - question-answering and 
        - table question-answering models

    - Deploys via `React` (frontend) and `Node.js,Express.js` (backend) for seamless user experience.

Installation & Setup

Prerequisites

- Python 3.x
- Docker & Kubernetes (Optional for Deployment)

Steps

1. Clone the repository
```
git clone https://github.com/techiescamp/mlops.git
cd mlops
```

2.  a virtual environment (Recommended)
```
python -m venv venv
source venv/bin/activate  # For macOS/Linux
venv\Scripts\activate     # For Windows
```

3. Install dependencies (if requirements.txt exists)
```
pip install -r requirements.txt
```

Go to directory on which project you needed and start working on it.

Future Enhancement

- Automated ML Pipelines using DVC & MLflow.
- Continuous Integration & Deployment (CI/CD) with GitHub Actions.
- Model Versioning and tracking experiments.
- Cloud Deployment with Docker & Kubernetes.
- Monitoring & Logging with Prometheus & Grafana.

License

This project is open-source and available under the MIT License. © www.techiescamp.com/