Plant Edibility Classifier

A machine learning application that classifies plant edibility from images using transfer learning.

⚠️ Safety Disclaimer: This is for educational purposes only. Do NOT rely on it for real-world edibility decisions.

Quick Start

1. Setup

python -m venv .venv
.venv\Scripts\activate  # Windows
pip install -r requirements.txt

2. Prepare Data

data/
  train/
    edible/
      image1.jpg
      ...
    not_edible/
      image2.jpg
      ...

3. Train

python -m src.train --data-dir data --epochs 10 --batch-size 16

4. Serve

uvicorn api.server:app --host 0.0.0.0 --port 8000

5. Use

• Web UI: http://localhost:8000/
• API: POST /predict with image file

Project Structure

plant-edibility/
├── src/           # Training code
├── api/           # FastAPI server
├── web/           # Frontend
├── models/        # Saved checkpoints
└── data/          # Your dataset

API Endpoints

• GET /health - Health check
• POST /predict - Upload image, get prediction
• GET / - Web interface

Training Options

• --arch resnet18|resnet50
• --freeze - Train only classifier head
• --epochs 10 - Number of training epochs
• --batch-size 16 - Batch size (reduce if OOM)

Environment Variables

• MODEL_CHECKPOINT - Path to model file (default: models/plant_edibility_resnet18.pth)