🐠 Fish Classifier API

AI-powered fish species classification using deep learning with EfficientNet-B0. This API can classify fish species and find similar fish in the database using image embeddings.

🌐 Live Demo

• API: https://fish-api-md7q.onrender.com
• API Documentation: https://fish-api-md7q.onrender.com/docs
• Health Check: https://fish-api-md7q.onrender.com/health

✨ Features

• ** Species Classification**: Classify fish species with confidence scores using EfficientNet-B0
• ** Similarity Search**: Find similar fish in the database using cosine similarity
• ** Web Interface**: User-friendly HTML interfaces for testing
• ** REST API**: FastAPI-based backend with automatic Swagger documentation
• ** Cloud Deployed**: Hosted on Render with automatic deployments

🏗️ Architecture

• Backend: FastAPI with Python 3.13
• Model: EfficientNet-B0 (pre-trained on ImageNet, fine-tuned for fish classification)
• Image Processing: PIL + torchvision transforms
• Similarity Search: Scikit-learn cosine similarity on image embeddings
• Deployment: Render cloud platform

🚀 API Endpoints

Method	Endpoint	Description
GET	/	Welcome message
GET	/health	Health check status
POST	/predict	Classify fish species
POST	/find-similar	Find similar fish (top_k parameter)

🖥️ Web Interface

Simple Test Interface

• Clean, minimal interface for basic testing
• Upload image → Get predictions
• Real-time results with confidence scores

Advanced Web Interface

• Modern UI with drag & drop support
• Tabbed interface for predictions and similarity search
• Progress indicators and error handling
• Responsive design

🛠️ Local Development

Prerequisites

• Python 3.9+
• pip or conda

Setup

# Clone the repository
git clone https://github.com/unnatii14/fish-classifier-api.git
cd fish-classifier-api

# Create virtual environment
python -m venv venv

# Activate virtual environment
# Windows:
venv\Scripts\activate
# macOS/Linux:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Run the server
uvicorn main:app --reload --host 127.0.0.1 --port 8000

Local Testing

1. Start the API server (port 8000)
2. Open simple_test_interface.html in your browser
3. For development, update API_BASE to http://127.0.0.1:8000

🔧 Model Details

• Architecture: EfficientNet-B0
• Input Size: 224×224 RGB images
• Preprocessing: ImageNet normalization
• Classes: Multiple fish species (see label_mapping.json)
• Similarity: Feature embeddings + cosine similarity

🔄 Deployment

This project is automatically deployed on Render with:

• Build Command: pip install -r requirements.txt
• Start Command: uvicorn main:app --host 0.0.0.0 --port $PORT
• Auto-deploy: Enabled on main branch commits

🙏 Acknowledgments

• EfficientNet architecture by Google Research
• FastAPI framework for the REST API
• Render platform for cloud hosting
• PyTorch and torchvision for deep learning infrastructure

Project Ecosystem

This repository is part of a three-tier fish identification system:

• Deep Learning Model - Core Deep learning model for fish species identification and Similarity Search.
• API Backend - RESTful API serving the Deep learning model
• AquaVision Frontend - Flutter Application for fish identification