Welcome! 🚀 This repo is part of my journey to dive deep into Deep Learning (DL) for computer vision. It’s a personal project to document my learning journey to implement DL architectures from scratch. I’ll update it as I progress.
This repository is a work in progress, where I:
- Explore and implement classic and SOTA DL architectures.
- Focus on understanding the inner workings of these models.
- Document my learning process and practical use cases.
-
Implemented Architectures:
- CNNs: ResNet, VGG, etc.
- Transformers: ViTs and variants.
- Lightweight models: MobileNet, EfficientNet.
- New approaches: Swin Transformers, ConvNeXt.
-
Flexible Training Framework:
- Pipelines for training and evaluation.
- Clone the repository:
git clone https://github.com/AhmedNasr7/vision-models cd vision-models - Install dependencies
pip install -r requirements.txt
This repo is for personal reference and growth. Feel free to explore or adapt it for your own learning!