README.md

CSE 555 Term Project (Computer Vision and Natural Language Processing)

Overview

This project is a multi-featured application focused on food image classification, variation detection, recipe recommendation, and reporting. It leverages deep learning and NLP techniques to provide a comprehensive toolkit for food-related data analysis and user interaction.

Features

• Image Classification: Classify food images using pre-trained models.
• Variation Detection: Detect variations in food items.
• Recipe Recommendation: Recommend recipes based on user input and image analysis.
• Report Generation: Generate reports based on classification and recommendation results.

Project Structure

PatternRec_Project_Group5/
├── assets/
│   ├── css/                # Stylesheets
│   ├── modelWeights/       # Pre-trained model weights (.pth)
│   └── nlp/                # NLP data and models (to be downloaded from google drive once the app runs)
├── config.py               # Configuration file
├── Scripts/                # Configuration file
│   ├── CV/                 # CV Training script
│   ├── NLP/                # NLP Training script
├── Home.py                 # Main entry point (possibly Streamlit or similar)
├── model/                  # Model code (classifier, search recipe)
├── pages/                  # App pages (image classification, variation detection, etc.)
├── utils/                  # Utility functions (layout, etc.)
├── sakenv/                 # Python virtual environment

Setup Instructions

1. Clone the repository:

   git clone <repo-url>
   cd PatternRec_Project_Group5
   

2. Create and activate the virtual environment: (Already included as sakenv/):

   source sakenv/bin/activate
   

3. Install dependencies:

   pip install -r requirements.txt
   

4. Run the application:
   • If using Streamlit:

     streamlit run Home.py
     

   • Or follow the instructions in Home.py.

Python Version

• Python 3.12.2

Notes

• Model weights are stored in the assets/ directory.
• NLP weights were quite large and are stored at CSE 555 Project Group 5
• Ensure you have the necessary permissions to access large files in assets/modelWeights/ and assets/nlp/.
• For best results, use the provided virtual environment and requirements file.