Collaborators

Amayas29 Ghiles-OUHENIA

Deep-learning-for-vision

This project focuses on leveraging deep learning techniques for computer vision tasks. The journey begins with a comprehensive introduction to neural networks, covering essential concepts like backpropagation, optimization methods, and a practical guide to using PyTorch.

Neural Network Fundamentals

Explore the core principles of neural networks, diving into the intricacies of backpropagation, optimization strategies, and a hands-on approach to implementing them using PyTorch.

Convolutional Neural Networks (CNN)

The project advances into Convolutional Neural Networks (CNNs), showcasing their effectiveness in image classification tasks. CNNs are applied to classify images from the MNIST dataset, and for the CIFAR-10 dataset, we adopt the VGG16 architecture to demonstrate their capabilities in more complex scenarios.

Transformers with Multi-Head Attention

In the final phase, the project explores the latest advancements with transformers, specifically using the multi-head attention mechanism. This cutting-edge architecture is applied to image classification tasks using the MNIST dataset, showcasing the adaptability of transformers in visual recognition.

Dive into the code and documentation to gain insights into each project phase, and witness the powerful applications of deep learning in the realm of computer vision. Contributions and feedback are greatly appreciated!