AI_Agent_Data_Explore

# 🤖 AI Agent Data Explore

A powerful, user-friendly **Streamlit** application for end-to-end data analysis and exploration—with AI-powered insights.

## ✨ Features
- **📂 Data Upload:** Easily upload CSV, Excel, or JSON files  
- 😈 **Model Setting:** Easily choose the right model  
- **🔍 Data Overview:** Instantly view data statistics, samples, and descriptive info  
- **⚙️ Analysis Target:** Define your analysis goals (*default:* “help me explore the data, show the insights”)  
- **🗺️ Analysis Plan:** Review the generated plan and give your feedback  
- **📊 Result Viewer:** Browse historical analysis reports & visualizations  
- **⭐ Evaluation:** Rate the quality and relevance of the analysis  

---

## 💡 UI Flow  
![image-20250419091330960](http://hexo.kygoho.win/upload/uploads/a99aa585-d745-4f59-b910-416a3ee4cf30.png)

[Visit the UI](https://app.uizard.io/p/65bf10b5)

## 🕸️ Agents Graph  
![image-20250417201155362](http://hexo.kygoho.win/upload/uploads/1a85a006-a525-4093-a344-4779e8a6159e.png)

---

## 🚀 Quick Start

### 📦 Local Installation

1. **Clone the Repo:**
   ```bash
   git clone https://github.com/nusduck/AI_Agent_Data_Explore.git
   ```

2. **Install dependencies:**
   ```bash
   pip install -r requirements.txt
   ```

3. **Set up API key:**
   Create and edit `.env` file:
   ```env
   OPENAI_API_KEY=Your_API_KEY
   LANGSMITH_API_KEY=Your_API_KEY
   LANGSMITH_TRACING=true
   LANGSMITH_ENDPOINT="https://api.smith.langchain.com"
   LANGSMITH_PROJECT=Your_LangSmith_project
   TAVILY_API_KEY=Your_API_KEY
   DEEPSEEK_API_KEY=Your_API_KEY
   GOOGLE_API_KEY=Your_API_KEY
   ```

4. **Run Streamlit app:**
   ```bash
   streamlit run streamlit_app.py
   ```

5. Open [http://localhost:8501](http://localhost:8501) in your browser.

---

### 🐳 Deploy with Docker Compose

> **Recommended for easy deployment and isolation**

1. **Create a `.env` file in the project root:**
   ```env
   OPENAI_API_KEY=Your_API_KEY
   LANGSMITH_API_KEY=Your_API_KEY
   LANGSM