YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Smart Visual Product Search Model


license: mit library_name: pytorch pipeline_tag: image-classification tags: - computer-vision - pytorch - efficientnet - image-classification - recommendation-system - retail-ai

Smart Visual Product Search - Product Classifier

This repository contains the custom trained EfficientNet-B0 product classification model used in the Smart Visual Product Search & Recommendation System.


Model Overview

The classifier predicts the product category from uploaded product images before retrieval and recommendation stages.

Architecture

  • EfficientNet-B0
  • Framework: PyTorch
  • Training Device: Kaggle GPU
  • Epochs: 30
  • Input Size: 224x224

Supported Categories

  • camera
  • headphones
  • keyboard
  • laptop
  • mobile
  • monitor
  • mouse
  • printer
  • speaker
  • watch

AI Pipeline

The complete recommendation pipeline:

  1. YOLO detects and crops the product region
  2. EfficientNet classifier predicts product category
  3. CLIP generates semantic embeddings
  4. FAISS retrieves visually similar products
  5. RAG recommendation layer generates intelligent recommendations

Performance

  • Validation Accuracy: 67.55%
  • Lightweight architecture for fast inference
  • Optimized for retail product retrieval

Files

File Description
best_product_classifier.pth Trained EfficientNet model
labels.txt Product category labels

Tech Stack

  • PyTorch
  • EfficientNet-B0
  • OpenCV
  • NumPy
  • Hugging Face Hub

Limitations

The current model supports selected retail product categories only.


Author

Aryan Gupta

B.Tech CSE (Cyber Security)
AI/ML & Data Analytics Enthusiast

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support