nailarais1/image-classifier-efficientnet

🌸 Flower Classification Model

📊 Model Info

Model: nailarais1/image-classifier-efficientnet
Author: Naila Rais
Task: Image Classification · 102 Flower Species

Quick Start

Installation

pip install torch torchvision pillow

Basic Usage

import torch
import torchvision.transforms as transforms
from PIL import Image

# Load model
checkpoint = torch.load('best_model.pth', map_location='cpu')
model = ...  # Your model architecture
model.load_state_dict(checkpoint['model_state_dict'])
model.eval()

# Predict flower
def predict_flower(image_path):
    transform = transforms.Compose([
        transforms.Resize(256),
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.485, 0.456, 0.406], 
                             std=[0.229, 0.224, 0.225])
    ])
    
    image = Image.open(image_path).convert('RGB')
    image_tensor = transform(image).unsqueeze(0)
    
    with torch.no_grad():
        outputs = model(image_tensor)
        _, predicted = torch.max(outputs, 1)
    
    return predicted.item()

# Get flower name
flower_id = predict_flower('your_flower.jpg')
flower_name = class_names[flower_id]  # Use class_config.json
print(f"Predicted: {flower_name}")

Model Info

• What it does: Identifies 102 different flower species
• Input: Flower images (224×224 pixels)
• Output: Flower name and confidence score
• Architecture: EfficientNet
• Training: 3 epochs on Oxford Flowers dataset

Example Results

• 🌹 Input: rose_image.jpg → Output: "rose" (98.2%)
• 🌻 Input: sunflower.jpg → Output: "sunflower" (95.7%)
• 🌷 Input: tulip.jpg → Output: "tulip" (92.3%)

Files Included

• best_model.pth - Trained model weights
• class_config.json - Flower names mapping
• config.json - Model configuration
• labels.txt - List of all flower names

Supported Flowers

102 species including:

• 🌹 Rose
• 🌻 Sunflower
• 🌷 Tulip
• 🌼 Daisy
• 💐 Lily
• 🏵️ Orchid
• 🌺 Hibiscus
• 🌸 Cherry Blossom
• And 94 more...

For Developers

# Get top-5 predictions
def top_k_predictions(image_path, k=5):
    # ... (implementation)
    return [
        {"flower": "rose", "confidence": 0.98},
        {"flower": "tulip", "confidence": 0.01},
        # ...
    ]

License

MIT License - Free for personal and commercial use ✅

Need Help?

• Model not loading? Check PyTorch version
• Wrong predictions? Use clear, centered flower images
• Other issues? Open a discussion on this repo

Download and start classifying flowers today! 🌸

Model by Naila Rais · Hosted on Hugging Face