D-Neeraja/car_vs_bike_classifier

🚗🏍️ Car vs Bike Image Classifier

This repository contains a Convolutional Neural Network (CNN) model trained to classify whether an image contains a car or a bike.

📊 Model Overview

• Input: RGB image resized to 128x128
• Output: Binary classification (0 = Bike, 1 = Car)
• Framework: TensorFlow / Keras
• Model Format: .h5 (HDF5)

🧠 Architecture Summary

• Conv2D → ReLU → MaxPooling
• Conv2D → ReLU → MaxPooling
• Flatten → Dense(128) → Dropout
• Dense(1) with sigmoid activation

✅ Performance

• Accuracy: ~96% on validation data
• Loss: ~0.12 (binary crossentropy)

🧪 Example Usage (Python)

from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing import image
import numpy as np

# Load model
model = load_model("car_vs_bike_model.h5")

# Load and preprocess image
img = image.load_img("test.jpg", target_size=(128, 128))
img_array = image.img_to_array(img) / 255.0
img_array = np.expand_dims(img_array, axis=0)

# Predict
prediction = model.predict(img_array)

if prediction > 0.5:
    print("Prediction: 🚗 Car")
else:
    print("Prediction: 🏍️ Bike")

📁 Files in This Repo

• car_vs_bike_model.h5 → Trained model file
• README.md → Project description & usage
• (Optional) example_images/ → Sample test images

📥 Download Instructions

Clone this repo and use Git LFS for large model files:

git lfs install
git clone https://huggingface.co/D-Neeraja/car_vs_bike_classifier

🧰 Requirements

• Python 3.7+
• TensorFlow ≥ 2.5
• NumPy

🙋‍♀️ Author

Neeraja Dakkata
📧 dakkataneeraja0031@gmail.com
🔗 GitHub: github.com/D-Neeraja

---

Feel free to use this model for educational purposes or integrate it into your own projects.