AI-Agent

Groq Data Extraction Flask App

Overview

A Flask-based web application that dynamically processes user queries with placeholders, replaces them with dataset entities, and retrieves data using the Groq API.

Features

• Dynamic Placeholder Replacement
• Groq API Integration
• User-Friendly Interface
• Secure API Key Management
• Robust Error Handling

Prerequisites

• Python 3.7+
• pip

Installation

1. Clone the Repository

   git clone https://github.com/yourusername/groq-data-extraction-flask.git
   cd groq-data-extraction-flask

2. Create & Activate Virtual Environment

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies

   pip install -r requirements.txt

Configuration

1. Set Environment Variables

   export GROQ_API_KEY='your_groq_api_key'  # Unix/Mac
   set GROQ_API_KEY=your_groq_api_key       # Windows

2. Prepare Dataset

   • Ensure your dataset is loaded into the df DataFrame in app.py.

Usage

1. Run the Application

   python app.py

2. Access via Browser

   • Navigate to http://127.0.0.1:5000/ and submit your query with placeholders like {Company}.

Example Query

• {Company}: Will be replaced with entries like "OpenAI", "Google", or "Microsoft".
• {Tourism}: Will be replaced with entries like "Eiffel Tower", "Great Wall", or "Statue of Liberty".
• {Vitamins}: Will be replaced with entries like "Vitamin C", "Vitamin D", or "Vitamin B12".

How It Works

When you submit the above query through the application:

1. Placeholder Detection: The app identifies all placeholders within curly braces {}.
2. Entity Replacement: Each placeholder is replaced with corresponding unique values from the dataset.
3. API Processing: The customized queries are sent to the Groq API to retrieve relevant data.
4. Results Display: Extracted information for each entity-placeholder combination is displayed on the results page.

Sample Output

For example, one of the processed queries might be:

The application will generate and display the extracted data based on this customized query.