A convolutional neural network (CNN) implementation for classifying images from the CIFAR-10 dataset using TensorFlow and Keras.
This project implements a CNN to classify images from the CIFAR-10 dataset, which contains 60,000 32x32 color images across 10 classes. The model achieves good accuracy while being relatively lightweight.
Key Components:
- Image preprocessing and normalization
- CNN architecture with convolutional and pooling layers
- Model training and evaluation
- Image classification inference
Technology Stack:
- Python 3.7+
- TensorFlow 2.x
- Keras API
- OpenCV for image processing
- Matplotlib for visualization
- NumPy for numerical operations
- CIFAR-10 dataset loading and preprocessing
- CNN model implementation
- Training with validation
- Model saving and loading
- Single image classification
- Image visualization
- Accuracy and loss metrics
- GPU acceleration support
- Simple inference pipeline
- Python 3.7 or later
- pip package manager
- (Optional) NVIDIA GPU with CUDA for accelerated training
- Clone the repository:
git clone https://github.com/yourusername/cifar10-classifier.git
cd cifar10-classifier