Here is a collection of projects I built while experimenting with existing models and libraries, sometimes building them completely from scratch. Since these are learning projects, they might not be perfect—better to check twice before using them! Contributions and ideas are always welcome.
| Project | Domain | Description |
|---|---|---|
| Digit Recognition | CV | A PyTorch-based Convolutional Neural Network (CNN) for MNIST handwritten digit classification, implementing custom architecture, data augmentation, and regularization techniques. |
| GPT2-From-Scratch | NLP | A PyTorch implementation of the GPT-2 architecture built for experimentation. Use it to train models on custom text or run inference using the original OpenAI weights. |
| Flickr8k Dataset | Multimodal | An organized distribution of the Flickr8k dataset containing images and text annotations. Structured for easy integration into image captioning and multimodal research projects. |