이 프로젝트는 Model Context Protocol (MCP)을 통해 MLflow에 자연적인 언어 인터페이스를 제공합니다. 일반 영어를 사용하여 MLflow 추적 서버를 쿼리 할 수 있으므로 기계 학습 실험 및 모델을 쉽게 관리하고 탐색 할 수 있습니다.

개요

ML 플로우 MCP 대리인은 2개의 주요 성분으로 이루어져 있습니다:

1. **MLflow MCP 서버 ** (mlflow_server.py): MLflow 추적 서버에 연결하고 모델 Context Protocol (MCP)을 통해 MLflow 기능을 노출합니다.

2. **MLflow MCP 클라이언트 ** (mlflow_client.py): 대화형 AI 보조를 사용하여 MLflow MCP Server와 상호 작용하는 자연적인 언어 인터페이스를 제공합니다.

특징

• **자연 언어 쿼리 **: 일반 영어에서 MLflow 추적 서버에 대한 질문
• ** 모델 레지스트리 탐험 **: 등록 된 모델에 대한 정보 -Experiment Tracking: 당신의 실험을 탐구하고 실행
• 시스템 정보: MLflow 환경에 대한 상태 및 메타 데이터를 가져옵니다.

필수품

• 파이썬 3.8+
• MLflow 서버 실행 (과태: http://localhost:8080)
• LLM의 OpenAI API 키

설치

Smithery를 통해 설치

MLflow MCP Server: Natural Language Interface for MLflow

This project provides a natural language interface to MLflow via the Model Context Protocol (MCP). It allows you to query your MLflow tracking server using plain English, making it easier to manage and explore your machine learning experiments and models.

Overview

MLflow MCP Agent consists of two main components:

1. MLflow MCP Server (mlflow_server.py): Connects to your MLflow tracking server and exposes MLflow functionality through the Model Context Protocol (MCP).

2. MLflow MCP Client (mlflow_client.py): Provides a natural language interface to interact with the MLflow MCP Server using a conversational AI assistant.

Features

• Natural Language Queries: Ask questions about your MLflow tracking server in plain English
• Model Registry Exploration: Get information about your registered models
• Experiment Tracking: List and explore your experiments and runs
• System Information: Get status and metadata about your MLflow environment

Prerequisites

• Python 3.8+
• MLflow server running (default: http://localhost:8080)
• OpenAI API key for the LLM

Installation

Installing via Smithery

To install MLflow Natural Language Interface Server for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install @iRahulPandey/mlflowMCPServer --client claude

Manual Installation

1. Clone this repository:

   git clone https://github.com/iRahulPandey/mlflowMCPServer.git
   cd mlflowMCPServer
   

2. Create a virtual environment:

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

3. Install the required packages:

   pip install mcp[cli] langchain-mcp-adapters langchain-openai langgraph mlflow
   

4. Set your OpenAI API key:

   export OPENAI_API_KEY=your_key_here
   

5. (Optional) Configure the MLflow tracking server URI:

   export MLFLOW_TRACKING_URI=http://localhost:8080
   

Usage

Starting the MCP Server

First, start the MLflow MCP server:

python mlflow_server.py

The server connects to your MLflow tracking server and exposes MLflow functionality via MCP.

Making Queries

Once the server is running, you can make natural language queries using the client:

python mlflow_client.py "What models do I have registered in MLflow?"

Example Queries:

• “Show me all registered models in MLflow”
• “List all my experiments”
• “Get details for the model named ‘iris-classifier’”
• “What’s the status of my MLflow server?”

Configuration

You can customize the behavior using environment variables:

• MLFLOW_TRACKING_URI: URI of your MLflow tracking server (default: http://localhost:8080)