https://github.com/chandru-21/llmops

This project(RAG) focuses on operationalizing LLMs by integrating OpenAI, MLflow, FastAPI, and RAGAS for evaluation. It allows users to deploy and manage LLMs, track model runs, and log evaluation metrics in MLflow. The project also features MLflow traces that logs all the user inputs ,responses ,retrieved contexts ,and other essential metrices.
https://github.com/chandru-21/llmops

fastapi genai llm llmops mlflow mlflow-tracking openai rag ragas

Last synced: 10 months ago
JSON representation

This project(RAG) focuses on operationalizing LLMs by integrating OpenAI, MLflow, FastAPI, and RAGAS for evaluation. It allows users to deploy and manage LLMs, track model runs, and log evaluation metrics in MLflow. The project also features MLflow traces that logs all the user inputs ,responses ,retrieved contexts ,and other essential metrices.

• Host: GitHub
• URL: https://github.com/chandru-21/llmops
• Owner: Chandru-21
• Created: 2024-10-17T09:02:21.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2024-10-28T07:30:43.000Z (over 1 year ago)
• Last Synced: 2024-10-31T08:06:28.371Z (over 1 year ago)
• Topics: fastapi, genai, llm, llmops, mlflow, mlflow-tracking, openai, rag, ragas
• Language: Python
• Homepage:
• Size: 2.55 MB
• Stars: 3
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Retrieval Augmented Generation (RAG)

## RAG LLMOps using OpenAI,MLflow,FastAPI and RAGAS(Evaluation)

Step I : Install all the requirements and execute the RAG_mlflow.py file

Now your LLM model run is registered in MLFlow with its evaluations metrices. 

![image](https://github.com/user-attachments/assets/d2114340-6f4b-44fb-a772-38e1d283a157)

RAGAS Evaluation metrics are logged in MLFlow, refer code for further details.

Step II : Execute the app.py (python app.py) in cmd navigate to the FastAPI UI in browser,

![image](https://github.com/user-attachments/assets/535549fc-3595-448c-b30c-2f4cd7a42638)

Click on try it out and enter your question

Step III : Model Monitoring/tracing

Navigate back to the MLflow UI and click on the 'Traces' tab. There, you'll find the question you asked in FastAPI is logged along with its response.

![image](https://github.com/user-attachments/assets/8f4037fb-7533-4ec1-bf20-70319418f252)

Click on the Request id to see more details about the run,

![image](https://github.com/user-attachments/assets/84fe7bdf-db3d-4872-ba70-233088442ef5)