Vibe Matcher - Fashion Recommendation System

A mini recommendation system that matches vibe queries to fashion products using OpenAI embeddings and cosine similarity.

Quick Start

1. Install dependencies:

pip install -r requirements.txt

2. Set your OpenAI API key:

# Windows CMD
set OPENAI_API_KEY=your-api-key-here

# Windows PowerShell
$env:OPENAI_API_KEY="your-api-key-here"

3. Launch Jupyter:

jupyter notebook vibe_matcher.ipynb

4. Run all cells to see the system in action!

What's Inside

• 8 mock fashion products with descriptions and vibe tags
• OpenAI embeddings (text-embedding-ada-002) for semantic matching
• Cosine similarity search to find top-3 matches
• 3 test queries with performance metrics
• Edge case handling with fallback suggestions
• Latency visualization and quality metrics

Features

✅ Semantic vibe matching beyond keywords
✅ Similarity scoring with quality thresholds
✅ Edge case handling (no matches, generic queries)
✅ Performance metrics and visualization
✅ Interactive testing section

Sample Queries

• "energetic urban chic"
• "cozy comfortable weekend vibes"
• "elegant sophisticated evening wear"

Try your own in the interactive section!