A modular deep learning framework for training and evaluating image classification models on datasets like CIFAR-10 and MNIST. Supports configurable CNN architectures, automated training, and performance visualization using Python and TensorFlow.

This project provides a modular deep learning framework for training and evaluating image classification models.

• Modular CNN architecture
• CIFAR-10 dataset support
• Training and evaluation scripts
• Accuracy visualization

datasets/ – dataset storage
models/ – deep learning model architectures
training/ – training scripts
utils/ – helper functions
notebook/ – experimentation notebooks

Python
TensorFlow
Keras
Google Colab

• Computer vision experiments
• Deep learning research
• Educational AI projects

pip install -r requirements.txt

python training/train.py

This project currently supports the CIFAR-10 dataset.

The CNN model achieves good classification accuracy on CIFAR-10.

• Add ResNet architecture
• Add MobileNet
• Add data augmentation

• Add ResNet and MobileNet models
• Support custom datasets
• Add data augmentation pipeline

MIT License